<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Brian Cheong on Medium]]></title>
        <description><![CDATA[Stories by Brian Cheong on Medium]]></description>
        <link>https://medium.com/@briancheong?source=rss-a5ecd526f4f8------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*wJwMfZS5IPaeoTDW.jpg</url>
            <title>Stories by Brian Cheong on Medium</title>
            <link>https://medium.com/@briancheong?source=rss-a5ecd526f4f8------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 00:38:07 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@briancheong/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[What can AI do for HR?]]></title>
            <link>https://briancheong.medium.com/what-can-ai-do-for-hr-7123d59157f5?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/7123d59157f5</guid>
            <category><![CDATA[humanresourse]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[hr]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Tue, 27 Feb 2024 11:21:01 GMT</pubDate>
            <atom:updated>2024-02-27T11:21:01.697Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*B-3tSAPpduYZgTI8-X7nnA.jpeg" /></figure><p>The heart of any organization is its human resources. From hiring to developing employees to evaluating their performance, human resources (HR) plays a crucial role in the continuous development of a company. On the surface, it may seem like a simple task. However, it’s actually one of the most complex and challenging jobs in the world. Since it’s mostly about dealing with people, it requires a wide range of experience and expertise. To do it well, stress management and excellent communication skills are essential.</p><p>When you talk to people in human resources, you’ll find that most of them are very interested in the efficiency gains that AI can bring to their work. They are also curious about how AI will change organizational structures and the role of HR within them. However, at the same time, many believe that HR is a human-to-human process that is difficult to automate and challenging to replace with AI. After talking to various HR professionals, and developing and testing AI agents, I have concluded that the process of HR management can be improved and made more efficient with AI. <strong>Improving the work of the human resources department means enhancing the operation and management of the company’s organization, which in turn leads to efficiency gains across the company.</strong></p><p>In this article, I’ll provide examples of how AI can be used in human resources and show how humans and AI can work together. All of the examples below are based on dummy data to illustrate the logic of applications across the business, and any references to specific individuals or companies are purely fictional. All demonstrations were created using Kompas AI (<a href="https://kompas.ai/">https://kompas.ai</a>)</p><h3><strong>1. Create a job posting (Job Description)</strong></h3><p>A job posting is an important first impression between an organization and a potential employee. A well-written job description clearly communicates the key tasks, responsibilities, required skills, and experience of the job, making it easy for candidates to determine if they are a good fit for the position. However, inaccurate or vague job descriptions can be confusing to candidates and may lead them to apply without a clear understanding of the job. While properly presented qualifications can help attract the right talent, a poorly written job posting that includes unnecessary requirements can discourage the right candidates from applying and reduce the efficiency of your hiring process.</p><p>Therefore, job postings should be carefully crafted. However, if your HR staff doesn’t have a good understanding of the job or lacks the necessary information, the job posting may be unclear or inadequately written. Getting the job description right can also be a challenge when the person in charge of hiring is inexperienced or when a company is expanding into a new area of business and needs to hire talent for unfamiliar roles.</p><p>In these situations, AI can be leveraged to quickly generate a basic job description (JD) and revise it based on feedback from the hiring manager, saving time while creating a more organized and cohesive job description.</p><p>Below is an example of a job posting for a Senior Marketer position for Kompas AI.</p><figure><img alt="AI can be leveraged to quickly generate a basic job description (JD)" src="https://cdn-images-1.medium.com/max/1024/1*G9SR3vOxwi540oVcRe6b3A.png" /></figure><p>The input is simple in this example; providing more detailed information can yield better results. In this way, AI can be used to quickly create a basic framework, which can then be built upon to craft a more detailed and sophisticated job description.</p><h3>2. Resume screening</h3><p>If you’ve written a compelling job posting, you’re likely to receive a lot of resumes from relevant candidates. If the job is very specialized or unattractive, the number of resumes received will be small, and you will have enough time to spend on each one. However, if the job is generic or attractive, you will receive a large number of resumes.</p><p>The purpose of resume screening is to quickly evaluate candidates based on their qualifications, experience, skills, etc., and select the most qualified ones for the interview stage. During this process, human resources personnel review candidates’ resumes to determine their fit with the job. It’s difficult to efficiently evaluate a large number of applicants; the right candidates can be overlooked, and the conscious or unconscious biases of screeners can lead to the rejection of qualified candidates or the selection of unsuitable ones.</p><p>This is where AI can be used to automate the screening process. By uploading resumes and inputting job descriptions, AI can assist in evaluating candidates more objectively. Unlike traditional automated resume screening tools that are limited to searching for keywords and terms, AI can take into account a deep understanding of the job requirements and the overall context of the resume to produce more efficient and valid results.</p><p>Below is a screen capture of the screening process. Based on the job description, the AI creates evaluation criteria, reviews each candidate’s resume according to these criteria, and presents the results.</p><figure><img alt="Based on the job description, the AI creates evaluation criteria, reviews each candidate’s resume according to these criteria, and presents the results." src="https://cdn-images-1.medium.com/max/800/1*sNImHFRPNAKq7DWzr9-y5g.gif" /></figure><h3><strong>3. Prepare for interviews</strong></h3><p>Once you’re done screening resumes, you need to set up a suitable interview for the candidate. Interviews are an important step in assessing a candidate’s skills and experience, as well as their fit with your company culture. Providing a positive interview experience can help your company build a good image and improve brand recognition.</p><p>Effective interviews require thorough preparation, structured questions, and clear evaluation criteria. However, if human resources personnel are interviewing candidates for various positions and do not have a good understanding of the job, they may not be able to ask appropriate and valid questions. Additionally, if they lack interviewing experience, the interview process can leave a negative impression on the candidate. An unprepared interviewer may miss important questions or fail to properly assess a candidate’s competencies, while conscious or unconscious biases can skew the interview process and lead to missing out on promising candidates.</p><p>To prevent this, HR departments work to train interviewers and set relevant standards. While AI can’t take over the whole interviewing process, it can aid interviewers in their preparation. The AI can generate a set of questions based on the job description and the candidate’s resume, which can be used as the foundation for a more meaningful interview. With a ready-made set of interview questions, interviewers can prepare for more in-depth discussions and convey to the candidate that they are well-prepared.</p><p>In the example, I entered the job description for a senior marketer and the resume of the candidate. AI was able to reference the details of the resume to create interview questions that are tailored to the candidate.</p><figure><img alt="AI was able to reference the details of the resume to create interview questions that are tailored to the candidate." src="https://cdn-images-1.medium.com/max/1024/1*xOHSwL0KgyyqdW-BJ-Vbcg.png" /></figure><h3>4. Set up candidate evaluation criteria</h3><p>In order for the hiring process to be effective, it’s important to set clear evaluation criteria early on. Having clear evaluation criteria minimizes personal bias and subjective judgment, ensuring that all candidates are evaluated consistently and fairly. The clarity of criteria facilitates better communication between interviewers and between line management and human resources, allowing for more refinement of future job postings and resume screening processes. Accurate evaluation criteria also help provide specific and useful feedback to candidates, which not only contributes to their improvement but also leaves a positive impression of the company.</p><p>For this reason, establishing clear hiring criteria and candidate evaluation metrics at the onset of the hiring process is essential. Once we have a job description and required qualifications, AI can assist in clarifying what needs to be evaluated during the hiring process. Based on the evaluation criteria suggested by the AI, we can quickly and easily finalize the scorecard by identifying what is missing or what needs to be re-ranked in terms of importance. Similarly, you can also create probationary evaluation criteria for new hires.</p><p>In the example below, AI outlined a set of criteria for evaluating candidates during the hiring process based on the Senior Marketer job description mentioned earlier. Using this as a foundation, you can modify and enhance the criteria by applying your own judgment and understanding of the role.</p><figure><img alt="In the example below, AI outlined a set of criteria for evaluating candidates during the hiring process based on the Senior Marketer job description mentioned earlier. Using this as a foundation, you can modify and enhance the criteria by applying your own judgment and understanding of the role." src="https://cdn-images-1.medium.com/max/1024/1*NJ6NnM0SnQhC3oymwZ37LA.png" /></figure><h3>5. Set OKR/KPIs</h3><p>One of the most important tasks in human resources management is company-wide performance management, where OKRs (Objectives and Key Results) and KPIs (Key Performance Indicators) are the most commonly used methods. OKRs and KPIs clarify the organization’s goals and provide specific directions to move toward them. They help teams and individuals focus on the key areas they need to improve.</p><p>Setting OKRs and KPIs can be more complex and challenging than it might seem due to the complexity of the business environment and the varying needs of different stakeholders. Properly setting and managing OKRs and KPIs requires time, people, and financial resources. When these resources are limited, effective execution can be difficult and even overwhelming for business leaders. Employees who are unfamiliar with OKRs/KPIs may misunderstand them or struggle to create appropriate metrics, leading to HR spending significant resources on training and administration.</p><p>When setting or reviewing OKRs/KPIs, AI can assist in this process. Based on the results, more detailed and accurate metrics can be set. In principle, OKRs should be written by each employee, setting their own goals. However, by using AI to create a first draft and then revising it based on guidelines, useful OKRs can be developed much faster.</p><p>In the example below, a set of company-wide OKRs would be presented, and then OKRs for the marketing team would be derived based on them. Next, the OKRs for each of the three members of the marketing team would be broken down, ensuring alignment with the broader objectives while tailoring them to individual roles and responsibilities.</p><figure><img alt="AI can assist in setting or reviewing OKR process" src="https://cdn-images-1.medium.com/max/1024/1*VvP_aNbY-LUF-CzmZOcDsA.png" /></figure><h3>6. Performance Review (Performance Evaluation)</h3><p>Performance reviews, or performance evaluations, are an essential part of organizational management. These reviews help employees develop and grow professionally by identifying their strengths and areas for improvement. With continuous feedback, employees gain a clear understanding of their performance and are motivated to achieve their goals. Recognition is a powerful motivator for employees.</p><p>However, preparing for performance reviews requires careful consideration. These evaluations must consider various metrics, goal achievement, personal characteristics, and input from other team members. If managers lack sufficient time for preparation, relevant experience, or fail to provide specific and constructive feedback during the evaluation process, it can have negative consequences. Additionally, delivering feedback, particularly negative feedback, can be stressful for managers.</p><p>To conduct effective performance reviews, HR and department heads should collaborate to ensure enough time is allocated for the evaluations and that communication methods are well-coordinated. However, this process can be highly individualized and demand significant resources.</p><p>This is where AI comes in to help us create an effective and valid feedback structure based on the information gathered. It’s important to remember that AI-provided guidelines are just a starting point. Department heads and leaders can use these guidelines, supplemented by their own experience and communication skills, to deliver more impactful feedback. AI-generated guidelines can help remove some of the guesswork from the process and make communication more efficient, particularly for new leaders who are still developing their review techniques.</p><figure><img alt="AI can help us create an effective and valid feedback structure for performance review." src="https://cdn-images-1.medium.com/max/1024/1*EQIfrSTUgR8gJ6uszbGUnA.png" /></figure><h3>7. Design a training program (Staff Training)</h3><p>Regular training for company members is crucial in fostering a culture of learning and development. When employees perceive that the company is invested in their growth, they tend to be more satisfied with their jobs and show greater loyalty to the organization. Employee training enhances efficiency and effectiveness, leading to improved productivity across the company. Furthermore, it can provide a compelling reason for employees to remain with the company long-term and help decrease turnover rates.</p><p>Nonetheless, developing suitable and effective training materials and programs tailored to an organization’s specific needs can be challenging. Training initiatives require time and financial investment to prepare and implement. While outsourcing to third-party specialists is an option, it can be costly and may not yield a program that aligns closely with the organization’s unique culture and objectives.</p><p>AI can help solve the challenges of preparing and delivering training programs. With the help of AI, HR departments can craft bespoke training programs and collaborate with department leaders to facilitate their delivery.</p><p>In the example below, I asked how to design a training program for new hires on time management and work prioritization. The AI provides suggestions for training content, methods, materials, and more. You can then refine and customize the training program to suit real-world training contexts, enhancing its relevance and effectiveness for the participants.</p><figure><img alt="I asked how to design a training program for new hires on time management and work prioritization. The AI provides suggestions for training content, methods, materials, and more" src="https://cdn-images-1.medium.com/max/1024/1*K5uXrIP8ymMDotzDjA7VMg.png" /></figure><h3>8. Human Resources Strategy</h3><p>HR strategy is a long-term plan designed to effectively manage and optimize human resources with the aim of achieving an organization’s business goals. It encompasses strategies for finding, hiring, and retaining the right talent over the long term. An HR strategy should be tailored to an organization’s specific context and revisited and adjusted periodically as market conditions change and the organization evolves</p><p>AI can help you quickly review your HR strategy to ensure that it is meeting your current business goals, identify gaps, and make improvements. This can significantly reduce the time and cost of strategy construction and review, allowing organizations to revisit and improve their strategies more frequently.</p><p>Below is AI’s analysis and suggestions for workforce redeployment and attrition plans as the company realigns its operations.</p><figure><img alt="AI’s analysis and suggestions for workforce redeployment and attrition plans as the company realigns its operations." src="https://cdn-images-1.medium.com/max/1024/1*Je9VJ0xECY5URB3_RlJ1Ig.png" /></figure><p>All the examples presented above were created using Kompas AI (<a href="https://kompas.ai/">https://kompas.ai</a> ). The AI agents mentioned are available directly through Kompas AI. Implementing similar functionality with ChatGPT is possible in a simpler manner, but it may not be easy to achieve the same level of results as the examples</p><p>The applications mentioned above are the result of leveraging AI not just as a knowledge tool, but as a reasoning engine based on a variety of information. Through this process, <strong>AI can assist people in making better decisions</strong>. While AI cannot completely replace human work, it can facilitate the completion of routine tasks more quickly and effectively. This enables us to devote more time and energy to more important matters and makes it easier to undertake tasks that were once out of reach.</p><h3>Finally, let’s take a look at what you need to keep in mind when applying AI to human resources</h3><p><strong>1) Avoid entering sensitive work data into services like ChatGPT, Gemini, or Claude.</strong><br>- Services like ChatGPT, Gemini (BARD), and Claude, which are easily accessible on the web, clearly state that the data you enter can be used for training their AI models. Therefore, <em>it is crucial not to share sensitive personal information (such as resumes), organizational goals, performance reviews, etc.</em> This precaution is vital not only for HR but for everyone in the company when integrating AI into their work.<br>- Alternatives include building your own large language model (LLM) infrastructure or using an AI solution developed for the workplace, like Kompas AI. These solutions are specifically designed to ensure that your data is not utilized to train AI models and can safely process sensitive materials.</p><p><strong>2) Don’t blindly trust AI</strong><br>- While AI can offer fresh insights or uncover knowledge we might not have considered, we must not treat the information provided by large language models (LLMs) as infallible or exhaustive. Particularly in cases where they are not finely calibrated, LLMs can produce “hallucinations” or generate misleading information. Therefore, it’s crucial to use AI with a discerning mindset, critically evaluating its output. <strong>Remember, AI is just a tool, and humans are the ones who make the final decisions.</strong></p><p><strong>Once we realize that AI can be of great help in the most “human” of fields, such as human resources, the potential for its application in other areas seems boundless. </strong>To learn more about AI’s impact on different professions and why its presence may not yet be widely felt, check out this article <strong>(</strong><a href="https://briancheong.medium.com/how-will-ai-change-our-jobs-ddab439616ab"><strong>How will AI change our jobs?</strong></a><strong>).</strong></p><p>In future articles, I will explore how AI can be utilized across various professions and share real-world examples. All demonstrations will be created using <a href="https://kompas.ai/">Kompas AI</a>. Currently, Kompas AI offers over 70 sophisticated AI agents that are ready for immediate integration into diverse professional roles. I invite all of you to experience Kompas AI firsthand and discover how AI can enhance their work.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7123d59157f5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How will AI change our jobs?]]></title>
            <link>https://briancheong.medium.com/how-will-ai-change-our-jobs-ddab439616ab?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/ddab439616ab</guid>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[gpt]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[productivity]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Tue, 30 Jan 2024 11:01:34 GMT</pubDate>
            <atom:updated>2024-01-30T11:01:34.686Z</atom:updated>
            <content:encoded><![CDATA[<p>I think it’s safe to say that the biggest buzzword of 2023 was generative AI. ChatGPT has more than 180 million subscribers and nearly 100 million weekly active users. Everyone, from the general public to the media, is talking about generative AI. This makes me consider trying it myself, but I’m at a loss for what questions to ask or how this technology could be applied to my job or company. After a brief test, the results don’t seem as impressive as I thought, and it seems like most people have concluded that AI still has a long way to go and have simply forgotten about it.</p><p>In this article, I’m going to talk about how AI agents are changing the business world and how they will affect our personal work and future, focusing on large language models (LLMs) like OpenAI’s GPT-4.</p><p>(Throughout the article, I use the terms AI, AI agent, generative AI, and LLM interchangeably for ease of understanding. Technically, they all have slightly different meanings, but I’m just using them in a way that’s easier to write and understand.)</p><h3>How does generative AI impact knowledge workers?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/700/0*FfJnCHuQ0gX-30FN.png" /></figure><p>The impact of AI on the knowledge worker job market is more significant and widespread than one might think. A recent study found that after the launch of ChatGPT and Midjourney, the number of jobs and earnings of copywriters and graphic designers on major online freelance platforms dropped significantly. This suggests that the products of generative AI are being subjected to direct competition with human-created output, directly impacting their jobs.</p><p>OpenAI estimates that new advances in AI pose the greatest risk to the highest-paid jobs, with high-wage workers being about three times more at risk than low-wage workers, according to its analysis. This indicates that AI could impact not only simple and repetitive jobs, but also jobs that require specialized knowledge.</p><p>The emergence of ChatGPT has significantly triggered discussions about AI strategy in many corporate boards, with AI adoption rates increasing even more dramatically. As generative AI solutions prove their value and become embedded in a wider range of business processes, their application is likely to expand further.</p><p>We need to recognize that AI is transforming jobs, not just replacing them. By automating repetitive tasks, AI enables employees to spend more time on strategic and creative work, which, in turn, improves productivity and efficiency, fueling business growth and creating new jobs. AI technologies are already fundamentally changing the business landscape, reshaping workforce structures and jobs across industries.</p><h3>Does generative AI really help productivity, and how?</h3><p>Github Copilot, a leading AI solution, helps software developers speed up the development process and reduce the likelihood of errors. According to recently published data, developers who used Copilot experienced more than a 70% improvement in code clarity and writing speed. AI assistance has been shown to be especially useful for novice developers, helping them learn and improve their coding skills faster. This shows the potential of AI to not only improve productivity and quality, but also serve as a learning tool.</p><p>A study of users of Microsoft’s AI solution, Copilot, found that: 1) users experienced an average of 27% speedup in tasks such as conducting meetings, writing emails, and creating documents. 2) In an evaluation by an independent panel of experts, emails written with Copilot were rated 18% better for clarity and 19% better for brevity. 3) AI-enabled users were also 27% faster at gathering and organizing information from multiple sources.</p><p>Harvard Business School recently studied the impact of GPT-4 on productivity and quality of work among employees at The Boston Consulting Group (BCG). The study found that BCG consultants assigned to use GPT-4 while working on a series of consulting tasks were more productive than their peers who did not use the tool. Specifically, the AI-assisted consultants performed their tasks 25% faster and handled 12% more workload, and their quality of work was rated 40% higher than their non-AI peers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3ga3puW0oTo0N-1Q.png" /></figure><p>One of the interesting findings of the study was that users in the bottom 50% of the evaluation saw a 43% improvement in performance by using AI tools, while the top users saw a 17% improvement. This means that all employees, regardless of skill level, benefited from utilizing AI, but especially those who were previously rated low. This suggests that AI can close the gap between people’s ability to perform tasks and improve the overall quality of work for everyone in the organization.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*w-Vu8vJWMatEBcLe.png" /></figure><p>The study also describes two groups of people who differed in the way they collaborated with AI. The first group, dubbed “cyborgs,” worked closely with the AI, constantly generating, reviewing, and refining answers, while the second group, dubbed “centaurs,” split the roles, focusing on their own areas of expertise and giving the AI subordinate, supportive tasks.</p><p>Finally, the study found that the role and effectiveness of AI did not appear to be the same in all areas of work. In particular, for some tasks that require a deeper understanding of the problem or more nuanced judgment in solving it, AI did not deliver the improvements noted above. This shows that while AI has the potential to improve productivity and quality in many areas, it cannot fully replace the deep cognitive abilities and expertise of humans.</p><h3>So why aren’t we there yet?</h3><p>Numerous studies and articles are already talking about a world transformed by AI. So why am I not taking advantage of AI? Why does no one around me seem to be using AI? What’s the problem?</p><p>As I talk to different people about using AI in their work, I realize that there is a huge gap in people’s understanding and ability to use AI. People have varying perceptions of the role, capabilities, and limitations of generative AI, as well as diverse approaches to its utilization and task delegation. This leads to a huge gap in the ability to utilize AI. If this trend continues, the disparity in output between individuals will only widen.</p><p>Many people still view LLMs like GPT as simple search engines or knowledge databases. As a result, it’s easy to underestimate the power of LLMs when they don’t have information about the latest events or can’t provide accurate answers about specific expertise.</p><p>In fact, we should understand LLM models like GPT as reasoning engines. Reasoning is the mental task of analyzing observed information or collected data to discover regularities or patterns, make predictions about future events, or draw conclusions for deeper understanding. This process is essential for making connections between information, coming up with solutions to complex problems, and generating new ideas. In a knowledge-based society, these reasoning skills are key to the daily work of a wide range of professionals, including scientists, engineers, doctors, lawyers, business analysts, and others. Studies have shown that in many fields, GPT-4 reasoning and decision-making skills are already higher than those of the average person. LLMs can play an important role in creativity, innovation, analysis, persuasion, and more.</p><h3>Large Language Models (LLMs) alone are not enough.</h3><p>Why does the LLM always seem to give a dumb (insufficient) answer whenever I use it? This is because LLMs lack both short-term and long-term memory and do not have appropriate reference materials or guidelines. LLMs have a limited context window, allowing them to focus only on brief information and moments. If you try to process too much information at once, you will not be able to process it properly and will lose your overall understanding. Without proper guidance, LLMs might start talking haphazardly, fabricate nonexistent facts, forget important details, or remember things incorrectly.</p><p>However, what makes LLMs potentially powerful is their ability to utilize a variety of tools and knowledge bases based on their reasoning capabilities. This allows LLMs to make better, more informed decisions and take action using a variety of tools. LLM performance improves dramatically. Of course, choosing the right data sources and the right tools are essential to making LLMs work for you. LLM is not just a search engine, it’s an inference engine.</p><h3>How can we make AI do its job better?</h3><p>The instructions we give to an AI are known as prompts. However, depending on which prompt you give it, the results you get can be completely different.</p><p>Let’s switch gears for a moment. When you first started your job as an entry-level employee or intern, you probably encountered many various mentors. Think back to when you first received your work assignment and started working. A good mentor will define the purpose of the task well, understand your abilities and experience, advise you on how to execute it, and let you know in advance where you might make mistakes. They also clearly communicate their expectations for specific deliverables. In this case, it’s much easier for a talented subordinate to understand the purpose of the work and do their best to deliver the required results. On the contrary, a bad mentor provides unclear definitions of tasks, lacks understanding of the other person, and asks for results in an abstract manner. Naturally, it’s hard to get a good result with such a work order. (The bad mentor might nitpick why the results are only this much…) In fact, collaboration with AI is very similar to this. Obviously, the good mentor here can write much better prompts and get better results. Of course, it would be ideal if you had a subordinate or an AI that can understand gibberish as if it were clear instructions, but that still seems difficult.”</p><p>1) People who know the role of AI as an inference engine and understand its capabilities and limitations can design more sophisticated prompts. 2) People who know what they need to input and what they want to get out of it will be able to write better prompts. In this sense, people who are more knowledgeable and experienced in their domain are more likely to write better prompts. 3) People who have experience working with diverse teams are generally more likely to use AI and chatbot platforms effectively. They are better at guiding AI to more accurate and useful responses. 4) The ability to structure your work also has an important impact on prompt design. People with a structured mindset are able to clearly set goals, design a course of action, and effectively communicate the information they need to the AI. When solving complex problems, this allows you to break down each part of the problem and craft prompts accordingly to get the most out of the AI’s capabilities.</p><p>Users can understand the AI’s responses in full context and, if necessary, ask for more information or follow up with more specific questions. This approach (similar to the cyborg type mentioned above) is crucial to getting better results from interactions with AI. You can play to the AI’s strengths in analyzing the data it provides, strategizing, and making decisions. This allows you to make the most of AI’s responses and work more efficiently and effectively. In fact, this ability is critical not only for AI-enabled work, but also for collaboration between human coworkers.</p><h3>Data Safety Issues</h3><p>ChatGPT offers good value for money. A free version is available, and with the Pro version, although there are some time limitations, you can continue to use GPT-4. Directly connecting to the API for the same amount of tasks would incur much higher costs. Why is ChatGPT cheaper?To borrow a famous saying from the industry, “If you’re not paying for the product, you are the product.” This means that if you’re not paying for it, your data is the product.</p><p>LLM companies explicitly state that their web-based products, such as ChatGPT and BARD, collect user conversation data and use it to train their models. These platforms rely on user-supplied conversation data to address data shortages and to collect a variety of use cases to improve their models. If you enter sensitive information into ChatGPT, it can be stored for as long as they want and used in the training process in any way they want. This means that there is a possibility of sensitive information being leaked. In fact, a recent case was reported where ChatGPT exposed email addresses used in the training process by including them in the output.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*T7g8oDl3qkkhR44I.png" /></figure><p>The above is probably a case where someone’s email was used for training and the results happened to look exactly the same. Of course, this also means that your work input into ChatGPT could be used as training data and appear in someone else’s output as shown above.</p><p>Alternatively, you can install the open-source LLM in an on-premise environment. This way, you can manage all of your data on your own servers. However, installing and securely maintaining such a system can be more expensive than you might think. Open source LLMs still have a significant performance gap over commercial models. While many benchmarks point to recent improvements in the performance of open source-based LLMs, it is very likely that the actual performance in real-world usage may be lower than what benchmark scores reflect. This is especially true in areas such as access to external knowledge and the use of tools. The gap is even more pronounced when prompting and receiving results in languages other than English. This gap is unlikely to close in the near term.</p><p>The next option is to use GPT as an API. OpenAI, Microsoft Azure, Google, and Anthropic all offer LLMs as APIs, and they all specify that the data is not used for training purposes. By sending only the data you need through the API, you don’t have to share the full context. This cannot be said to be completely safe. However, it’s worth remembering that many organizations had similar concerns when AWS, GCP, and other cloud services were first introduced. Eventually, the performance, cost, and overall utility outweighed the initial concerns. If you’re already using AWS, GCP, Slack, Notion, Google Docs, etc. for your business, the additional concerns about using LLM APIs may be unnecessary. (There’s no significant difference)</p><p>In conclusion, the most effective and accessible way to use LLM at this point is to develop your own solution utilizing the APIs in GPT, Claude, and Gemini, or use a service that indirectly utilizes LLM APIs, such as Kompas AI. If you choose to develop your own, be prepared to deal with a variety of technical challenges, such as using Retrieval-Augmented Generation (RAG), system prompts, memory, and tools. To get the most out of an LLM, you’ll need to do a lot of fine-tuning (often very labor-intensive), which is not well understood across many disciplines and requires a lot of expertise (experience through lots of trial and error).</p><h3>So how does this apply to our work?</h3><p>Several studies have shown that LLMs excel in the following areas: 1) creativity and innovation, 2) analyzing, persuading, and managing, 3) providing initial feedback on work, and 4) helping with decision-making and providing a second opinion. For example, a writer or designer can use an LLM to generate multiple versions of a text or design concept and then select or refine the best one. Business analysts and data scientists can also use LLM to extract insights from data and analyze them further to make better, more informed decisions faster.</p><p>As mentioned above, LLM is a reasoning engine, not a search engine. By ensuring that your prompts reflect your clear goals, how you want to proceed, and the expected outputs, and by connecting the right knowledge and tools, you can perform complex and varied tasks and get high-quality results that you can use immediately in your work. Here are a few use cases</p><h3>Here are a few examples of use cases (a bit long)</h3><p><strong>1) Swiftly creates straightforward market competition reports.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*88c9LhAEyXRHmElZ.png" /></figure><p><strong>2) Conducts strategic analysis with various strategic frameworks.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*9nRHdUFDUHfDnHAP.png" /></figure><p><strong>3) Analyzes the potential market impact based on hypothetical scenarios.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*LXlMB1PzrEDQZlbm.png" /></figure><p><strong>4) Processes and visualizes data, albeit with some hiccups, but the results turned out well</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*og5l4YkTXFqwq-cy.png" /></figure><p>In addition to the aforementioned tasks, an AI agent:</p><ul><li>Offers valuable insights for business decision-making processes.</li><li>Assists with complex research.</li><li>Processes and analyzes complicated data.</li><li>Proposes creative plans and strategies.</li><li>Supports both coding and debugging.</li><li>Enhances user experiences with UX writing assistance.</li><li>Screens a multitude of resumes automatically.</li><li>Writes and distributes OKRs and KPIs.</li><li>Analyzes data from diverse documents to compile reports.</li><li>Organizes complex thoughts and aids in setting priorities.</li></ul><p>These examples highlight the broad range of tasks an AI agent can handle, among many other possibilities.</p><p>By the way, most of these AI agent examples here cannot be properly executed in ChatGPT. First of all, you need to be able to write the various prompts yourself, and if you try to run more complex tasks, ChatGPT will often reject them (due to resource constraints), and even if it does, it will most likely time out at some point and crash.</p><p>Running a sophisticated AI agent requires unrestricted operational resources and an infrastructure that can handle multiple iterations. One must also overcome the 60-second execution constraint of traditional code interpreters. As for search engines, Google’s results are significantly better than Bing’s. (In non-English-speaking regions, Bing’s search results are almost disastrous.) To implement these features, you need to develop your own infrastructure for LLM operations, set up your own AI agents, or use a service like Kompas AI. The aforementioned outputs were all generated using AI agents based on GPT and Claude through Kompas AI.</p><p>I keep mentioning Kompas AI because I am developing it. To give a brief introduction to the product:</p><ul><li>Kompas AI offers well-designed AI agents for various purposes, allowing users to easily and quickly derive excellent results.</li><li>Kompas AI provides AI agents that solve much more complex and difficult problems based on Google integration, very high iteration counts, support for various LLM models, and a powerful Advanced Code Interpreter.</li><li>There are also features for team-based users, so you can create customized AI agents and share them among team members.</li><li>Best of all, Kompas AI has strict policies on data governance and privacy, ensuring that no data is used for training. Users have full control over their data.</li></ul><p>If you’re interested, sign up for a free trial at <a href="https://kompas.ai/">https://kompas.ai</a>.</p><h3>What if you’re not using AI yet?</h3><p>According to a recent report from Salesforce, 28% of respondents stated they are already using AI in their work. More than half of them, 51%, expect AI skills to increase their job satisfaction, and 44% expect to earn a higher salary than their colleagues without AI skills.</p><p>At the same time, however, many companies have not yet provided clear guidelines on the potential and ways to apply AI in work. Among those who use ChatGPT for work, 55% report using AI services without formal permission from their company. It seems that business leaders still see AI as a distant and incomplete future technology. AI has shown significant advancements throughout 2023, and more changes are expected in 2024. Even now, we need to understand the capabilities and limitations of AI tools and have appropriate usage policies in place. We need to ensure that our members are able to fully utilize new technologies to increase productivity while keeping their data secure.</p><p>Finally, to quote World Economic Forum economist Richard Baldwin, “AI won’t take your job, but someone who uses AI will” I would add to this, “<strong>AI won’t ruin your company, but a company that effectively uses AI will</strong>”.</p><p>It’s never too late to start, even when you think it is. Compared to the overall changes we are going to experience, we are still at the very beginning of change.</p><p><em>There may be inaccuracies or omissions in this content due to my limited experience and knowledge. Please point it out so I can improve.</em></p><p>Reference:<br>1. <a href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4573321&amp;ref=blog.kompas.ai">https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4573321&amp;ref=blog.kompas.ai</a><br>2. <a href="https://www.salesforce.com/news/stories/ai-at-work-research/?ref=blog.kompas.ai">https://www.salesforce.com/news/stories/ai-at-work-research/?ref=blog.kompas.ai</a><br>3. <a href="https://github.blog/2022-09-07-research-quantifying-github-copilots-impact-on-developer-productivity-and-happiness/?ref=blog.kompas.ai">https://github.blog/2022-09-07-research-quantifying-github-copilots-impact-on-developer-productivity-and-happiness/?ref=blog.kompas.ai</a><br>4. <a href="https://www.microsoft.com/en-us/worklab/work-trend-index/copilots-earliest-users-teach-us-about-generative-ai-at-work?ref=blog.kompas.ai">https://www.microsoft.com/en-us/worklab/work-trend-index/copilots-earliest-users-teach-us-about-generative-ai-at-work?ref=blog.kompas.ai</a><br>5. <a href="https://www.ft.com/content/6f016bc2-8924-11da-94a6-0000779e2340?ref=blog.kompas.ai">https://www.ft.com/content/6f016bc2-8924-11da-94a6-0000779e2340</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ddab439616ab" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Next Plan]]></title>
            <link>https://briancheong.medium.com/next-plan-85168beacfdf?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/85168beacfdf</guid>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Fri, 22 Sep 2023 12:49:45 GMT</pubDate>
            <atom:updated>2023-09-22T12:49:45.531Z</atom:updated>
            <content:encoded><![CDATA[<p>We have previously shared with the community detailed information about Upbit’s delisting decision, as well as our project’s statement on the matter. The validity of our statement, Upbit’s decision, and the fairness of that decision are all separate issues.</p><p>With the sharp decline in prices and the spread of uncertainty in the market, many are pessimistic about the future of our project. Consequently, both MARO and ACN have seen significant price drops, inflicting deep wounds on our entire community. We would like to once again express our profound regret over this situation.</p><p>Over the years, as we developed our blockchain network and products on top of it, we deeply reflected on the growth mechanics and potential of the Maro blockchain and the Acorn protocol. Through this, <a href="https://medium.com/@ttcfoundation/maro-ecosystem-a-development-update-6e640ffe51ba">we were able to confirm certain outcomes and gain a more profound belief in the potential of a blockchain-based economic system and its capabilities as a value exchange platform.</a> With this belief, we continue to invest in product development and network enhancement.</p><p>Since the end of last year, we have taken a keen interest in generative AI technologies, contemplating their potential impact on labor productivity. We firmly believe that generative AI will provide heightened productivity for workers globally, leading to a rapid reshaping of the labor market. Aspects like data labeling and derived methodologies like RLHF (Reinforcement Learning from Human Feedback) will be continuously required in model development and fine-tuning, with demands continuously rising. In this context, we see the Acorn protocol logic playing a pivotal role. We’ve been persistently testing this technology since early this year and plan to launch a related product this October. Alongside this, we intend to focus our resources on developing practical market solutions through collaboration with generative AI solution developers, data labeling firms, and other external partners.</p><p>The governance token of the Acorn protocol, the OAK token, has been undergoing internal testing since last year, waiting for the optimal launch timing. We were anticipating that the introduction of Acorn protocol’s DAO and the OAK token would be significant milestones for the entire project when the market reaches adequate liquidity and scale. However, in light of recent events, it seems we may need to postpone this. Beyond the planned OAK, we aim to refine the staking mechanism of ACN and introduce a more sustainable solution.</p><p>We are fully aware that due to the price decline of Maro, node operators might find their rewards insufficient. These operators play a crucial role in network maintenance and management. We’re deliberating on supplementary solutions to ensure a safer and more stable network, including adjusting gas fees and other incentives. An increase in network users and usage through the Acorn protocol will augment the revenue for Maro network operators and those who voted for them. We strive to bolster the project’s sustainability through these measures.</p><p>Lastly, to ensure the continuous growth and expansion of the project, we will intensify our efforts to negotiate with domestic and international exchanges and expand liquidity. We plan to collaborate with various investors and partners to build a better ecosystem. Much of this information is being prepared confidentially, but we will introduce it when it becomes publicly shareable.</p><p>The market is harsh, and many have diverse opinions about us. Moreover, this situation has indirectly spawned many other problems. Nevertheless, we are committed to overcoming these challenges, restructuring our business, and relentlessly striving for liquidity expansion and growth.</p><p>Survival holds the only meaning, and by enduring, we can earn another chance to prove ourselves. For that opportunity, we will keep moving forward. Separately, we once again apologize and express our regret for the series of events.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=85168beacfdf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Our 2021 and 2022]]></title>
            <link>https://briancheong.medium.com/our-2021-and-2022-e80fd9dca79d?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/e80fd9dca79d</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[data-labeling]]></category>
            <category><![CDATA[self-driving]]></category>
            <category><![CDATA[omar]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Sun, 30 Jan 2022 08:18:52 GMT</pubDate>
            <atom:updated>2022-01-30T08:18:52.080Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gfCXB5bxsjNCP1aJoZXPSw.jpeg" /></figure><p>As we were wrapping up projects that took off during the 4th quarter of 2021, 2022 began in a quite hectic manner. January also flew by as well once we were done evaluating our 2021 performance and concluded solidifying our 2022 plans. In this article, I will briefly share our assessment of 2021 and plans for 2022.</p><h3><strong>2021</strong></h3><p>2021 was the year where we built our business with a much dedicated mind and time horizon that visioned far beyond short term accomplishments.</p><p>We launched “ByteBridge”, a web based SaaS that allows machine learning developers to seamlessly set up their data labeling projects and rapidly acquire the results. And, with WorkBox, we created a platform where anyone with a smartphone can be enabled to participate in even the most complicated type of data labeling, such as 3D Point Cloud annotations. Beyond these accomplishments, we also built and refined various products that are needed to manage and operate the overall process.</p><p>As a result, we were able to provide most forms of data labeling services related to images and videos, on ByteBrige. Our world’s first mobile based 3D Point Cloud viewer and labeling tool especially allowed us to show that even the most complicated data labeling needs could be met in a short amount of time by specifying, distributing, and automating the process with a mass worker base.</p><p>In 2021 we officially established a sales team and progressed the project by communicating with various clients in diverse fields from all over the globe. This gave us a better understanding of the market in terms of its competition dynamics and its exact needs; which in turn, advanced our confidence in the value our services will provide to the market.</p><p>Since Q3 of 2021, our team has been focusing on projects that are related to 3D Point Cloud. Further accumulating data labeling technologies and working with/on projects based on 3D Sensor Fusion has especially been our focus. By constantly building and improving our solutions and stacking up experience with numerous clients working on diversified projects, we expect that we will be partnering with many more bigger and prominent corporations very soon. Currently, we are working on, or planning to work on, many different projects with a number of automotive companies and autonomous driving solution development companies. If our clients agree to disclosures regarding our partnership, we will share these details.</p><p>Amid the progresses we have made, it is true that our resources were allocated mostly on discovering and testing projects alongside establishing a solid business model. This inevitably deprioritized the development of blockchain related software, infrastructures, and services. The decisions we have made the past year were long term strategies to better grow our business, but we do admit that these were quite unrelated from the hot topics of the year such as DeFi and NFTs.</p><h3><strong>2022</strong></h3><p>In the first half of 2022, we will be focusing on scaling up and expanding on projects that we have been building since the second half of 2021. We will advance the entire process so that more projects can progress faster, more work can be done more seamlessly, and more people can join in.</p><p>In tandem with such developments, we will strengthen our sales and marketing efforts in many more countries to acquire more clients and achieve stronger partnerships. Also, to allow services on the Maro Blockchain to have more integration with actual businesses, we will expand the contact points of these services. As the autonomous driving industry has exploded last year and that we expect this trend to continue throughout the next few years, we are determined to grab the opportunities that will present itself.</p><p>The macro economics and the financial/crypto markets of 2022 seems like it has a higher chance to take a turn for the worse compared to 2021. To be decoupled from these economic crises, we strive to promptly establish strong business and growth models.</p><p>When we look back into the past several years, I believe bear markets bestowed much greater opportunities. To better seize those opportunities much more easily, we will focus on building and strengthening our fundamentals.</p><p>At least by the mid term of 2022, we plan to resume working on expanding the Maro ecosystem and focus on developing Maro 2.0. The overall goals for these developments still remain (1) lowering the entry barrier for new users and (2) increasing the liquidity within the ecosystem. In addition to these objectives, the team is internally researching ways to evolve Tigris Protocol and Acorn Protocol into a DAO like system. We hope to build a system where more people can participate in managing/operating services, earn more rewards for their contribution, and eventually establish an environment where both the holders and products succeed. To assist these goals, helpful functions such as liquidity mining, liquidity voting, and more will be supported.</p><p>As the curtain rises for full-scale business developments, the development for automated integration between Acorn Protocol and actual businesses will take place as well. Once the related functions are better realized, partners are lined up, and legal matters are taken care of, we will share the details of our plans and achievements.</p><p>When we first began the Maro project, I think in a small sense the team blindly focused on developing and realizing the features that we had in mind. However, since 2021, we have been better focusing our energy in defining the essence of our business, growth, and value realization. We will always keep in mind to have a crystal clear long term vision and base our decisions off of that rather than waste our energies following short term market trends.</p><p>We strongly believe that if Maro can be integrated with well-built businesses to create value and benefit even more users, the ecosystem will grow naturally. We will steadily yet aggressively strive to achieve our long term goals to accomplish all that I shared above.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e80fd9dca79d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Realizing online mass collaboration and valuable data creation through blockchain]]></title>
            <link>https://briancheong.medium.com/realizing-online-mass-collaboration-and-valuable-data-creation-through-blockchain-a12870f61149?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/a12870f61149</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[data-labeling]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[mass-collaboration]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Wed, 02 Dec 2020 08:16:46 GMT</pubDate>
            <atom:updated>2020-12-02T08:16:46.909Z</atom:updated>
            <content:encoded><![CDATA[<p>As per<a href="https://medium.com/@briancheong/covid19-and-the-accelerating-change-of-digital-economy-7632bb47c98b"> the article written in the first half of this year</a> , our team enables large-scale collaboration through blockchain, generates meaningful data in large quantities, and creates value through this process all the while contemplating ways to build a better ecosystem. As a result of these efforts, ponderance, and numerous tested hypotheses, we have beta launched <a href="http://workbox.org/">WorkBox</a>, a blockchain-based remote working solution, and <a href="http://bytebridge.io/">ByteBridge.io</a>, a data annotation solution for machine learning developers.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FkRokLM2OJ-M%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DkRokLM2OJ-M&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FkRokLM2OJ-M%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/9a075e034f2e5b52384ffcc8f44ee292/href">https://medium.com/media/9a075e034f2e5b52384ffcc8f44ee292/href</a></iframe><p>To summarize the somewhat lengthy past article, <a href="http://workbox.org/">Workbox</a> divides various complex workloads into simpler tasks and utilizes its own consensus mechanism to allow numerous users to promptly work on such tasks via a mobile app. ByteBridge let ML developers easily start, manage and complete the whole process with efficiency. The consensus mechanism also allows the quick verification of submitted tasks. In the overall process, all tasks are settled and rewarded in real-time through tokens, while blockchain <strong>1) micropayments</strong> (no matter how small the amount is), <strong>2) instant payments </strong>(convenient and fast), and <strong>3) smart contracts</strong> (once conditions are met the transactions are made) enable <strong>mass collaboration across various countries on a trustless basis.</strong></p><p>In this article, I would like to introduce <strong>projects that have been conducted with real clients </strong>over the past few weeks after the hypothesis has been established and the product developed, and further elaborate on how this whole process has progressed.</p><h3><strong>Classification of mass image contents</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u9iiVIu3FmbhYZOEKvbt0g.jpeg" /></figure><p>This task targets to classify images suitable for data training by classifying only images that meet all eight conditions from a pool of hundreds of thousands of unclassified images and records the classification results separately. From the client’s point of view, when the amount of data they possess is vast, it requires a lot of time and resources to hire personnel to perform classification work. But with our solution, it is possible to classify large amounts of data in a short amount of time. Looking at the actual progress results, <strong>it can be seen that sorting hundreds of thousands of images takes a few hours while a million units can usually be processed within 24 hours.</strong> In a similar sense, there was a case where a specific hashtag of Instagram and all contents within a specific account were classified and sentiment analysis of the posting content was performed, and the result was extracted very quickly and accurately.</p><h3><strong>Highway license plate boxing</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*Bf0DsEVOMlFDdTGvRsoDew.jpeg" /></figure><p>Tasks labeling the number of vehicles on the road, the location and contents of license plates based on the video data collected by surveillance cameras for collecting traffic information. The provided video is converted into images frame-by-frame, labeled according to each frame, and the data is combined. Although we experienced some issues along the way, such as the way the cars are arranged and the difficulty of the mission changing depending on the number of cars on a screen (whether there is a traffic jam or not). As the project progressed, it was possible to significantly improve classification and location recognition method of objects, and the participation barrier (difficulty) was reduced by further subdividing the images. Labeling of training data for autonomous driving is done in a similar format, just the boxing types, 2D BOX, 3D BOX, 3D Cloundpoints, etc., are slightly different.</p><h3><strong>Portrait photo collection of ethnically Black users</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/540/1*PqLyWvohJ5hKrJfWG2OXXw.jpeg" /></figure><p>For companies located in Asia, it is relatively difficult to obtain facial data of different ethnicities such as from Europe and Africa. In this case, you have no choice but to purchase the training data you want at a cost, which is why we initiated the portrait photo collection project of ethnically Black (African) users. Since there were few existing users of Acorn Box in Africa, we started promoting Acorn Box and Workbox in collaboration with the African blockchain community. According to about 190 photograph categories requested by our client (day/night, male/female, indoor/room, etc.), users would take photos and submit them, while other users would review the submitted photos to see if they met the conditions. Once a user’s submitted photo was evaluated and verified, the user was rewarded in tokens right away. While conducting this project, we found that the CDN service we use completely blocks out certain countries and it was also quite clear there are huge differences between countries in terms of the widely used mobile phone models and their subsequent gaps in image quality. Nonetheless, through this project, we were able to confirm once again that even with people on the other side of the globe, we can generate, verify, and trade original data without existing trust relationships. During this project, as a result, we were able to collect tens of thousands of photographs of ethnically Black users.</p><h3><strong>Multi language voice collection</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*5o8eEMHa-tq1je8B-3TsTQ.jpeg" /></figure><p>A project that collects voice data necessary to support a wider variety of languages, such as smart speaker builders, automatic customer consultation solution providers, and companies that analyze the situation and emotions of speakers. According to the manuscript prepared in advance, people from various countries and regions make voice recordings (“lower the volume”, “turn off the lights”, from simple commands to simulating conversations in a specific situation, etc.), and other users evaluate the recordings and texts. Immediately after development, we built an audio database for Vietnamese, English, and Bengali as an internal project and confirmed that it was possible to collect massive voice content in a short time. In the near future, building a German/French audio database will be underway with real clients. Similar to how we conducted the image collection project with the African community, we also plan to secure German-speaking and French-speaking users and initiate the aforementioned project with our clients.</p><p>There were other big and small projects worth mentioning, but only the notable ones were introduced in this article. Although the above projects were summarized and delivered with a simple tone-and-manner, in fact, our team has worked tirelessly on each project solving various bugs and improving upon unforeseen problems. However, the good news is that most of the problems we have encountered have been solved. As we further work on more projects with our clients, we continue to improve the accuracy of our solutions and the speed of the overall process.</p><p><strong>I believe, and look forward to, that the combination of WorkBox and ByteBridge that we are creating will be the first commercial solution with blockchain technology that can enable numerous people to collaborate at the same time, allow everyone to create valuable data, and even provide a legitimate reward system for all stakeholders.</strong> In this ecosystem of ours, the fast and efficient <a href="http://ma.ro/">Maro Blockchain </a>is faithfully playing its role and <a href="http://acorn.eco/">Acorn Protocol</a> is allowing more quality data to be generated at a faster rate all the while making the entire process transparent and fair.</p><p>The hypothesis established during the coronavirus pandemic in the first half of 2020 is now leading to actual business opportunities, and the proof of its possibility is also starting to visualize. <strong>In the short term, we work to change the data labeling industry; in the midterm, we challenge ourselves to realize online mass collaboration; and, in the long term, we thrive to establish a new economic system based on a blockchain network.</strong> And, these efforts to accomplish our many goals continue to this day, even at this very moment.</p><p>Maro Blockchain <a href="http://ma.ro/">http://ma.ro</a><br>Acorn Protocol <a href="http://acorn.eco/">http://acorn.eco</a><br>WorkBox <a href="http://workbox.org/">http://workbox.org</a><br>AcornBox <a href="http://acornbox.org/">http://acornbox.org</a><br>ByteBridge <a href="http://bytebridge.io/">http://bytebridge.io</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a12870f61149" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[我们目前生活在一个分割的时代]]></title>
            <link>https://briancheong.medium.com/%E6%88%91%E4%BB%AC%E7%9B%AE%E5%89%8D%E7%94%9F%E6%B4%BB%E5%9C%A8%E4%B8%80%E4%B8%AA%E5%88%86%E5%89%B2%E7%9A%84%E6%97%B6%E4%BB%A3-3b40bdb0ce73?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/3b40bdb0ce73</guid>
            <category><![CDATA[繁荣]]></category>
            <category><![CDATA[合作]]></category>
            <category><![CDATA[分割]]></category>
            <category><![CDATA[区块链]]></category>
            <category><![CDATA[疫情]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Wed, 16 Sep 2020 06:47:06 GMT</pubDate>
            <atom:updated>2020-09-16T06:47:06.640Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8ncOWw5-aZddHevij_1d2w.jpeg" /></figure><p>我们目前生活在一个分割的时代。</p><p>由于新冠疫情，国际间的交流被迫中止，国家之间形成对立局势。全球政治气氛正从繁荣的全球化走向民族主义和地方保护主义。人们把“他们”和“我们”区分开来，并把两者分成不同的派系。全球社会正在恶化成为一个压迫和惠顾他人也不再有负罪感的阶段。连接和合作的机会正在逐步减少，而民族和地方利己主义正在超越全球的基础价值准则。随着人和人之间的关系越来越分裂，将会有更多人的人选择通过压迫别人来保护自己。</p><p>但是，从历史的角度来看， 人类的成就和繁荣，都是通过人和人之间合作交流完成的。二战之后，借助全球化的力量，我们以极快的速度发现探索世界、取得了无数新的发明创造。虽然有时候，这些成果的归属划分难免存在分歧。但不可否认的是，由此创造了整个人类历史的迅速发展。人类、国家和企业的进步和成功不是零和博弈的结果，而是稳定和合作增长的共同结果。</p><p>我们联系越紧密，开展各种协同合作的机会就越多。现代互联网科技的持续发展帮助我们更有效的参与到大规模的群体合作中。新兴科技突破了物理、地域、和联属的限制，使我们能够更顺畅地直接参与到全球经济活动中。通过互联网网络经济，所有的参与者都能创造更大价值，并且所有人的贡献都能得到公开透明的认可。这将为所有人带来积极的结果，并提高全球的社会效率和效用。</p><p>到目前为止，价值一直是由企业或政府决定的。许多障碍阻碍了个人最大化他们的价值和扩展他们的梦想。但是从现在起，个人的价值不再由某个实体来决定，而是通过同样参与网络的同辈来决定。不管怎样，我们都在创造价值，不管价值有多小，每个人的价值创造都应该被认可和尊重。而我们个人的价值能通过市场得到更好的发现和尊重。</p><p>我们梦想有这样一个世界，在这个世界上，没有人被排除在自己所拥有的机会之外，没有人被排除在为世界繁荣作出贡献的机会之外。我们希望更多的人能够克服阻挡这些机会的障碍。最后，我们希望更多的人可以达成更大的个人成就，并过上幸福的生活。</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3b40bdb0ce73" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[We are currently living an era of division]]></title>
            <link>https://briancheong.medium.com/we-are-currently-living-an-era-of-division-60a5fcaf16a2?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/60a5fcaf16a2</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[connection]]></category>
            <category><![CDATA[prosperity]]></category>
            <category><![CDATA[cooperation]]></category>
            <category><![CDATA[divison]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Wed, 16 Sep 2020 06:35:42 GMT</pubDate>
            <atom:updated>2020-09-16T06:35:42.239Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8ncOWw5-aZddHevij_1d2w.jpeg" /></figure><p>We are currently living an era of division.</p><p>The pandemic has severed national connections and forced the people to exhibit fear-based hostility to out groups. The global political atmosphere is changing from the prosperous ways of globalization to nationalism and protectionism. People are distinguishing “them” from “us” and dividing the two into separate sides. The global society is deteriorating to a state where oppressing and patronizing others are no longer subject to guilt. Opportunities for connection and cooperation are dwindling whereas national and local egoism is overtaking the fundamental notion of the world. As the division among people worsens, more and more people tend to think that protection can be sought through the oppression of others.</p><p>However, if we look back into the history of humanity, we can clearly see that human greatness, our most recognizable achievements, and opportunities for prosperity came from times of connection and cooperation. Post WWII, we have accomplished many great things at an exponential rate through the power of globalization. It is also true that the division of such fruition was not always fair and just. Nonetheless, it cannot be denied that there were indeed great accomplishments. The advancement and success of mankind, nations, and corporations are not the result of a zero sum game but are a shared outcome of a steady and cooperative growth.</p><p>The more we are connected the more opportunities we have for various synergistic cooperation. The continuous advancement of processing technology and network technology allows people to take part in mass collaboration more efficiently. New technologies support world wide users to directly participate in the global economy as they harness the power of individuals by overthrowing physical, location, and affiliation limits. All participants of the connected economy network can be transparently recognized for their contributions and create greater values online. This will bring about positive outcomes for all and increase social efficiency and utility throughout the globe.</p><p>Thus far, value has been determined by corporations or governments. Many obstacles have hindered individuals from maximizing their value and expanding upon their dreams. From now on out, the value of an individual will not be decided by a certain entity but through peers of the participated network. One way or another we are creating value and no matter how small that maybe, everyone’s value creation should be recognized and respected as is. Our value will be better discovered and well respected through the market.</p><p>We dream of a world where no one is excluded from the opportunities to use what they have and contribute to the world for prosperity. We hope more people will be able to prevail over obstacles that hinder such opportunities. And finally, we wish that more people can accomplish more and live a prosperous life.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=60a5fcaf16a2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What to focus on in 2020]]></title>
            <link>https://briancheong.medium.com/what-to-focus-on-in-2020-43bddc2d2d25?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/43bddc2d2d25</guid>
            <category><![CDATA[wallet]]></category>
            <category><![CDATA[marketing]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[ttc]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Thu, 27 Aug 2020 06:36:31 GMT</pubDate>
            <atom:updated>2020-08-27T06:36:31.872Z</atom:updated>
            <content:encoded><![CDATA[<p>What matters most for us?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1jr98bMv0hY6-3UTq_1PYw.png" /></figure><p>Since the very beginning, the year 2020 has dramatically changed our lives with the COVID-19 crisis. The results were a rapid change in the overall economic structure, various supply of liquidity to prevent economic collapse, fluctuations in numerous asset prices, “Uniswap” bound DeFi concept pumping and dumping, and appearance of subsequent scams. The environment has changed drastically, and the resulting noises are loud and undeniable.</p><p><strong>Changes in the external environment allowed us to think more deeply about what values ​​are important and what we should focus on</strong> (conversely what not to focus on); and as a result, based on certain principles, resources are now much more effectively concentrated. For the remainder of the year, we will continue to work on and advance upon various aspects based on such principles. In this article, I would like to share those principles and the direction we will take from now on.</p><h3><strong>Strengthened Marketing</strong></h3><p>Compared to what we have built and accomplished so far, we have been on the passive side when it comes to packaging and presenting our achievements to the outside world. The decision was made under the judgment that the overall market environment is immature for a full-scale expansion and that we need more time to fruitfully impact the market. However, we concluded that marketing and branding activities could not be executed suddenly at any one moment. Hence, from the later months of this year, we will prepare and proceed more systematically in terms of marketing.</p><ol><li>We have been organizing and redefining what we have been doing so far and what to do in the future.</li><li>We have been working on clarifying the vision and mission we pursue. Based on the principles established through such contemplation, we plan to conduct more effective and consistent branding and marketing campaigns.</li><li>We plan to focus more on the product story and concentrate on expanding our performance based on existing products.</li><li>We plan to conduct marketing more actively than we have done so far and promote various collaborations and partnerships more flexibly.</li></ol><h3><strong>User Expansion</strong></h3><p>We believe that the product/project direction needs to be more clearly aligned with the user’s intentions and behavior. We plan to focus our work on enabling the product to have a clearer purpose and acquire more users that fit the product’s purpose.</p><ol><li>For the time being, all channels regarding user acquisition will be integrated into Acorn Box. With Acorn Box, we are concentrating on making ACN more accessible, usable, and acceptable to the mass population. Acorn Box has been developed to be more intuitive and seamless to use than any other existing wallet, but we think it still has room for improvement.</li><li>Through the collaboration of WorkBox and ByteBridge, more people will participate in economic activities simply through their phones and be transparently rewarded for their work. In addition to the currently focused machine learning and artificial intelligence industries, we will devote resources to porting more diverse economic systems online.</li><li>As the use of the TTC Blockchain increased and the scale of the service itself has expanded recently, a bottleneck is occurring in smart contract usage and computation processing during specific time periods. In this regard, the development team is currently restructuring tata’s backend and Acorn Protocol’s reward computation related systems.</li></ol><h3><strong>Linkage Enhancement Among Products</strong></h3><p>The linkage between various products developed in the TTC Ecosystem is still weak, leading to rather uncomfortable user experience. Accordingly, our team will develop and improve various solutions that can strengthen the organic linkage between projects.</p><ol><li>InstantSwap and stablecoins will be used more actively as a product-to-product linkage solution. The overall network usage will increase and transactions between InstantSwap and TDEX will also become more active.</li><li>As part of the above, the total usage of Tigris Protocol is expected to continually rise. The users that are in the center of our focus when we plan and develop products are actual users of the service, not professional asset investors. We aim to create sustainable growth through our products and introduce new financial solutions to more and more people. In the process of such growth, Tigris Protocol’s stability will be substantially improved.</li><li>The movement and propagation of assets across the TTC network will become much easier. Solutions that can be seamlessly linked with more products will also be provided. During such advancements, we expect a smoother yet increased influx of new users.</li></ol><h3><strong>Increased Network Liquidity</strong></h3><p>The overall lack of liquidity in the network is one of the major causes of weak linkage among products. Hence, increasing liquidity is also an aspect we target to improve.</p><ol><li>InstantSwap will connect with Acorn Box and TTC Connect to help link not just only services within the ecosystem with each other, but also with external blockchain networks and exchanges. This will make it much easier for users to access a richer source of liquidity, whereas it will also allow more people to use the service simply. This series of processes will make the transition between assets easier and improve connectivity outside the TTC Network. Of course, in order to make this process even smoother, we plan to connect with diverse exchanges in the different markets.</li><li>ByteBridge’s service payment is first converted to CUSD via Tigris Protocol, which in turn is converted to ACN through InstantSwap and distributed to WorkBox users. This process, which automatically links InstantSwap and TDEX, is expected to further expand overall liquidity.</li></ol><p>Various external noises can make it daunting to get a grip on ourselves. Nonetheless, we are striving to remain composed even by the noisiest of noises, not fall into mannerism, and continue pursuing what we originally intended to achieve. Our team will focus on our vision and principles, and strive to create meaningful results.</p><p><strong>TTC will connect more people, help them collaborate, and produce socially admirable outcomes. We will develop solutions that are easy-to-use and accessible to billions. We believe this will eventually lead to user expansion, increased network usage, and ecosystem vitalization.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=43bddc2d2d25" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introducing ByteBridge.io]]></title>
            <link>https://briancheong.medium.com/introducing-bytebridge-io-9c02e40d1a83?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/9c02e40d1a83</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[traning-data]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[data-labeling]]></category>
            <category><![CDATA[data-collection]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Wed, 19 Aug 2020 07:58:11 GMT</pubDate>
            <atom:updated>2020-08-19T07:58:11.472Z</atom:updated>
            <content:encoded><![CDATA[<p>The unprecedented pandemic had me contemplate various aspects of today’s global society from diverse perspectives. I would like to introduce the result of such ponderings as it crystallized into a product that has surfaced into the world. In this article, I will highlight the meaning and concept of data required for machine learning, the current industry situation, and how we will explain and overcome the problems that we are currently facing.</p><h3><strong>ByteBridge.io</strong></h3><p>ByteBridge.io is a solution that makes data collection and processing more efficient and cheaper than anywhere else. It provides a service that is more than 90% cheaper than Google and any other Silicon Valley competitors, shows 10 times or more rapid data processing speed, and provides such automated services with transparency. The ByteBridge.io solution can be seamlessly used by anyone who would like to collect and process data.</p><h3><strong>Data Collection and Processing Work</strong></h3><p>When it comes to machine learning (especially supervised learning), data is the fuel. Just as cars cannot operate without fuel, machine learning cannot be realized without data. And, to implement high-level machine learning, you require substantial amounts of data. The task of collecting the data necessary for training is called data collection. For a specific purpose, it is necessary to collect data suitable for that very purpose, but since the algorithm cannot (yet) properly classify and designate those data, manual assistance from numerous people is required in the process of collection and classification. Hence, it is through manpower that the data used for machine learning is classified and processed into a form that can be used by the machine. Ironically, cutting-edge technology such as machine learning requires a lot of human input.</p><p>With increasing interest in machine learning and artificial intelligence around the world, demand for data collection and processing has also exploded. In particular, there are already numerous data collection and processing companies in China, where the related industry began to develop early, and many companies in India are supplying processed data to companies in the United States and Europe with their bases in Silicon Valley.</p><p>Except for large companies like Google, where the amount of data itself is abundant, most companies lack the proper data quantity required for machine learning. So, most companies have to purchase essential processed data. For example, to train a facial recognition solution, facial photo data of various races from various countries are required. Although large global companies already have a lot of data collected in their database, most startups are having difficulty collecting such needed data. Because systematic data collection is essential for better solutions, many businesses are bound to focus on finding a better way to collect and process data at a lower cost for efficient machine learning.</p><h3><strong>Current Industry Situation</strong></h3><p>As previously explained, data collection and processing requires a lot of manpower. Therefore, most data processing companies have offices in China and India, where labor costs are relatively cheap, and operate data factories in the form of direct management of personnel for processing data or re-outsources the work. The whole process of operation is actually very similar to how a workforce office operates.</p><p>When a data consumer (A) orders a data-related task (labeling or collection) to a data provider (B), the data provider 1) distributes the work to its own team (C1) located in various countries or 2) subcontracts to an outsourcing company (C2) that can handle the related work. And, this subcontractor (C1, C2) again recruits people (D) to engage in the required work. The subcontractors then train the recruits regarding the related work in a data factory similar to an internet cafe or office filled with computers to intensively process the requested data. After the subcontractors © complete the first data inspection, the data is passed on to the data provider (B), where the data provider goes through its own data inspection process once again to provide the results to the data consumer (A).</p><p>Unlike how the internet industry works, this process resembles the way traditional production lines and suppliers operate. The inefficiencies incurred in this process are very high, resulting in longer processing times and higher overall costs. Due to information asymmetry and costs incurred in the secondary and tertiary distribution stages, companies are forced to pay high costs, and the profits left to those who actually collect and process data are less.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*RoovE4gk7cBFSIS0" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/961/0*MN8U7DMOt-zlKBsb" /></figure><p>(Google also provides data labeling services through data factories that it operates or collaborates with, and even these data factories are closed down amid the COVID-19 crisis)</p><h3><strong>Opportunity</strong></h3><p>I believe that we can achieve great innovation in this field based on our product development capabilities and underlying blockchain-based technology. 1) Blockchain’s cross-border micro-payments and smart contracts make it easy to build a foundation of trust and eliminate intermediate steps. 2) The foundation’s consensus mechanism for data verification allows numerous people to collaborate on various tasks at the same time and enables rapid production of accurate results. 3) We are able to implement the entire process on a network to make the process transparent to our customers and allow for a clear prediction of the time for completion and total cost required to process the data, greatly improving the efficiency of the overall logistics.</p><p>I will expand upon how each participant in the process can benefit from our model,</p><h3><strong>Data Provider</strong></h3><p>Data providers/processors can easily participate in data labeling and data collection tasks through a mobile application (Acorn Box), regardless of time and place. The algorithms allow participants to receive the rewards for their work in real-time (usually within minutes) as soon as their work results are verified at high speed.</p><p>The mobile service development capabilities we have accumulated thus far provide a solution that enables participants to reliably produce better results regardless of their mobile phone model. The consensus algorithm, which is inspired by blockchain technology, enables various people to simultaneously participate in a single task and efficiently generate accurate results. Blockchain technology transparently and fairly evaluates the contributions of participants through smart contracts, and supports fast and inexpensive payments.</p><p>We provide opportunities for new income generation during these harsh times of a pandemic where the real economy is threatened and many jobs are lost around the world,</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FkRokLM2OJ-M%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DkRokLM2OJ-M&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FkRokLM2OJ-M%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/314276fe87cf3811c0db9be03e578dab/href">https://medium.com/media/314276fe87cf3811c0db9be03e578dab/href</a></iframe><h3><strong>Data Consumer</strong></h3><p>From the point of view of a data consumer (company or developer), now you can transparently check the data collection and processing process, which has been in a black box, and receive the desired data in an overwhelmingly fast and inexpensive way.</p><p>As numerous data providers are able to produce the desired results much more efficiently, data processing times that normally take more than a week are reduced to hours.</p><p>Since most of the data processing is completed on the network, the data consumer can keep track of the overall progress, speed, time to completion, and cost required in real-time, thereby improving work efficiency and risk management. Developers can upload data and download processed results through ByteBridge’s dashboard. Via the provided API, all processes such as data transmission, processing, and download can be easily connected with existing programs as well.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/961/0*g0eLun_1QMH0-0oC" /></figure><p>(ByteBridge Dashboard Example, It can be seen from above that 7,200 of the total 8,000 images have been processed, and 10,800 boxing has been complete. To date, the total cost has been incurred at $75.6, indicating that the total cost estimate is expected to be settled at $84)</p><p>Data processing costs, which have been determined through negotiations, are determined more transparently and can be significantly lowered, reducing inefficiencies and time waste that occurs in the negotiation process. This will allow developers to spend more time on what matters more.</p><p>Compared to existing Western companies, the service can be provided at a maximum of 90% lower rate whereas compared to companies in China and India, the service can be provided at a 50% cheaper price. And, despite the cheaper rate, the processing speed can improved by more than 10 times. If you sign up for ByteBridge.io now, you will be provided enough credit to process about 8,500 image boxings, so we encourage you to sign up and test it out.</p><p>Website <a href="https://translate.google.com/translate?hl=en&amp;prev=_t&amp;sl=auto&amp;tl=en&amp;u=http://www.bytebridge.io"><strong>http://www.bytebridge.io</strong></a></p><h3><strong>Go Big</strong></h3><p>As intermediate management companies, especially intermediate management personnel, are replaced with blockchain and algorithms, we can eliminate unnecessary costs and process much larger quantities of data at a fraction of the price compared to existing companies. In other words, we can offer more services to a wide range of clients with more data processing manpower at a lower cost. As a result, we expect to be able to scale our services at a rapid pace, and we expect this flywheel effect to further lower the cost per service.</p><p>We can reduce the burden of intermediary communication as most of the processes are automated, hence, we expect to be able to expand and apply the service to other countries more seamlessly. We are currently concocting marketing campaigns and sales strategies in various countries to be able to offer much more competitive pricing and improved efficiency in any country in the world.</p><h3><strong>Finally</strong></h3><p>According to Peter Thiel’s “Zero to One”, a new service needs to be improved by at least 10 times to make a meaningful impact on the market to dominate it. As we prepared and developed a new service, we immersed ourselves in service development under the hypothesis that our service could deliver that level of results. And, during the actual verification tests, we were able to confirm that our hypothesis was correct. Now, as the service fundamentals are near completion, we will begin marketing and sales both domestically and internationally. If there is an opportunity for collaboration, please feel free to recommend our service. Furthermore, if you are interested in Growth Hacking or B2B Sales, please <a href="/service/mailto:brian@bytebridge.io">contact us</a> directly.</p><p>Fyi, ByteBridge.io recently introduced on Yahoo Finance and Bloomberg</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*seOyI2QqKpp4DJ7EU5zNVQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CsyC18ngF9AI4hDbzM7Frg.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9c02e40d1a83" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[COVID19 and the Accelerating Change of Digital Economy]]></title>
            <link>https://briancheong.medium.com/covid19-and-the-accelerating-change-of-digital-economy-7632bb47c98b?source=rss-a5ecd526f4f8------2</link>
            <guid isPermaLink="false">https://medium.com/p/7632bb47c98b</guid>
            <category><![CDATA[data-collection]]></category>
            <category><![CDATA[data-labeling]]></category>
            <category><![CDATA[digital-economy]]></category>
            <category><![CDATA[dataannotations]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Brian Cheong]]></dc:creator>
            <pubDate>Wed, 15 Jul 2020 10:52:52 GMT</pubDate>
            <atom:updated>2020-07-15T10:52:52.600Z</atom:updated>
            <content:encoded><![CDATA[<p>Building a Global network economy system</p><p>With the advent of the Internet, our economic system has continually expanded online and endlessly reorganized. This change, which has been going on for about 30 years, has already made enough rapid changes for most people around the world. And, the COVID19 crisis which worsened at the beginning of 2020 is playing a huge role in accelerating those very changes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*neHvO59fsLgkk4YB_3MRtA.png" /><figcaption>Telecommuting experienced by many people (Source: 1password blog)</figcaption></figure><p>The normalization of the non-face-to-face interactions and the acceleration of the digital economy transition have been major changes over the past few months, most of which we could firsthand see and experience with the real economy. These changes are altering our society to a whole new look and seem to be irreversible on a certain level. Along with the world that has changed, the rules of the games throughout industries are also rapidly changing. Some industries quickly shattered amid this change, while others exploded. The labor structure reorganized and the direction of investment changed as well. Many are worried about the recession, and investors, who have no clear direction, ceased all progressive actions and are spectating from the sidelines.</p><p><strong>Blockchain and the Global Network Economy</strong></p><p>Year 2020 is a difficult time for most people, but it is also a good time to consider the nature of business. It could be considered ironic, but in the sense of crisis, we can feel our survival instincts grow stronger and think more clearly. With great changes in the global economy and drastic advancement of economic structure present, we must be able to quickly sense where we are headed and how the world will further change into; and, be able to lead the change with in-depth contemplation.</p><p>The concept of labor and asset is transforming quickly, and I think blockchain will act as an accelerator of these changes. We have already seen these changes happen during the current COVID19 outbreak. We experienced that many tasks could be promptly replaced with remote or online work, more efficiently and effectively than we expected. As the share of online in our lives continues to grow, the interest in data collection, accumulation, and utilization will increase evermore rapidly; also, the trend toward assetization of data will progress at a growing pace as well. Now, we need to invest much swiftly and boldly and lead the flow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*PL1pX1Bl6kA_MmoCgIeqGw.jpeg" /><figcaption>Could this be the famous (..) fourth industrial revolution? (Source: aerotek)</figcaption></figure><p>As I first planned and developed the TTC ecosystem, the ongoing question that I have in my mind is, “How can the mass create more meaningful and valuable data and get reasonably compensated for their contribution?” This question is continuing as a deeper contemplation on how we can integrate the value that comes from the acts of creating (labor) and acts of rewarding such actions (wage), the economic system, into a protocol.</p><p>In a network economy system, value is mostly created through the act of generating, accumulating, and processing data. The act of creating data on social networks is one such act of value creation. In this sense, the proposition of the Acorn Protocol, “Help create data on various platforms, and transparently provide evaluation and compensation for that data,” can be applied equally to extended concepts. Through this concept extension, we intend to expand and develop the Acorn Protocol beyond social networks to a protocol that can embrace all value-creating behaviors made online.</p><p><strong>Redefining Acorn Protocol…and WorkBox</strong></p><p>I believe that Acorn protocol has the potential to build a global network economy. We are building a network economy system that anyone can participate from anywhere in the world and can be transparently rewarded for their contributions. Anyone who has access to the Internet can be directly connected to the global economy and participate in its various economic activities. We make it easy for anyone to create valuable data that has an economic impact.</p><p>To realize this vision, we needed to redefine the concept of the Acorn Protocol. We are rebalancing our priorities by efficiently adjusting the opportunities we captured and the resources we have. Introducing WorkBox, a service we are vigorously focusing on and preparing in Q2 and Q3 of 2020.</p><p>We are preparing the official launch of WorBox, as elaborated, an online value creation platform. WorkBox is a service that anyone can create value through the Internet and receive real-time rewards as much as the value they created. The service will be in the form of a DAPP that is listed on and accessible through AcornBox (<a href="https://acornbox.acn.eco/wallet/download">a mobile wallet available on Apple Appstore and Google Play</a>). The first industry that can participate in the global network economy and benefit from WorkBox will have deep roots in supervised machine learning. In other words, the AI industry/machine learning field which are closely related to data labeling and data collection will have the most demand at the initial stage. Machine learning in the AI ​​industry, which is definitely in the spotlight recently, requires continuous training through numerous data, i.e. supervised learning; and this process requires constant cycling through data collection and data labeling. However, even to this day, these tasks are still being completed through the manual labor of many people. We expect that WorkBox will be able to create an unprecedented highly efficient system by allowing the mass to easily participate in data labeling and data collection.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ilFERpxEz1Q5cXoir5-c6A.png" /><figcaption>Data Annotation</figcaption></figure><p>With WorkBox, based on the experience we have accumulated, 1) anyone can participate as much as they want, 2) we can guarantee the accuracy of the results through our consensus mechanism, 3) reward the work done transparently and equitably through smart contracts, and 4) reward contributors in real-time regardless of national borders and time zones.</p><p>We believe that by utilizing the strengths of micro-payments, instant payments, and smart contracts of the TTC Blockchain, we can effectively build an economic system with higher efficiency and drastically lower management costs. As a result of internal testing so far, it is expected that WorkBox can maintain the compensation level of the service providers (workers) while being able to deliver results at a 70% lower cost and 7 times faster than the current average market statistics. Client companies can get their desired results much faster at a lower price, and suppliers can easily “work,” any time anywhere while receiving more abundant rewards.</p><p>The industry that can benefit through WorkBox will initially be those in need of data labeling and data collection, the AI industry/machine learning. But, we will gradually expand the fields to various businesses that have the need for data to be verified online. We will continue to port larger industries to our platform and gradually grow the overall economy.</p><p><strong>The Flywheel of Ecosystem Expansion</strong></p><p>Our vision and mission will not end with simply connecting the mass service providers and demand companies. Our flywheel logic, which we introduced in the previous articles, “The Flywheel Effect: TTC &amp; ACN” and “TTC + DeFi”, will continue to take effect and be further strengthened.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/778/1*33LUoNX3hegspzpYwYVlTw.png" /></figure><p>First of all, we expect that when the online value creation platform and the decentralized financial system are linked, we will be able to exert a stronger influence. Those who are rewarded through WorkBox can easily stake and receive interest through Tigris Protocol. In line with the launch of WorkBox, we will launch a stablecoin staking service so that workers can receive stable interest while preserving the value of their assets. Users in developing countries still have limited access to banks, so we expect the staking service can serve as a kind of bank. The stablecoin staking service will significantly lower the participation threshold for decentralized financial systems for people outside the existing blockchain industry. Of course, through this process, users’ assets will stay on the TTC network much longer.</p><p>The linkage between various services within the TTC network, especially the linkage/combination of each service with the decentralized financial platform, will become more frequent. To cope with this progress, 1) we are working on a merger on Tigris Protocol with the official launch of TDEX, a decentralized exchange. And at the same time, 2) the Instant Swap service will be launched for more natural linkage between various services. Instant Swap is expected to seamlessly supply liquidity between various products in the TTC ecosystem and satisfy the differing demands of both developers and users.</p><p>When the services that are being prepared are launched and connected, it becomes easier for the many users to create value on our network, earn money, save or invest. This will provide the foundation for even more DAPPs of all types and purposes to join the ecosystem and target growth. The flywheel of ecosystem expansion will still operate as it was fundamentally designed, and the influx of capital and employers will gradually become stronger.</p><p><strong>WorkBox</strong></p><p>The WorkBox service is currently being beta tested and is aiming for its official launch in August. The WorkBox taskforce is continuously surveying the demand for companies and services. Starting from May, we took over work from actual client companies and began testing the service. Through our own consensus mechanism and token economy, we will keep on improving the service efficiency at a rapid pace, continue to provide fast and cheaper services to more and more demanding companies, stably secure customers, and distribute larger quantities of work and rewards to more workers. To lower the entry barrier for both client companies and data workers, the Instant Swap is also currently being beta tested since early July</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/936/0*Y_EQbZA2cNVrMhTs.png" /><figcaption>You can participate in data labeling on workbox</figcaption></figure><p><strong>Finally</strong></p><p>I hope that our platform can help anyone across borders and time zones collaborate together and create greater values. With Acorn Protocol, we hope that more people will have further opportunities and more companies will be able to improve or solve their in-hand problems more efficiently. Based on this synergistic relation, we will create an even larger ecosystem; companies from various countries and industries will participate as consumers, and suppliers from various countries will be able to create value. Wouldn’t it be great to look forward to the arrival of a new global network economic system that will be built upon such economic logic?</p><p>For more information, you can refer to</p><ul><li><a href="http://www.ttc.eco/">TTC - A Blockchain Fit for Mass Adoption and a Sustainable Token Economy Model</a></li><li><a href="http://www.acorn.eco/">Acorn Protocol - A Decentralized And Token Incentivized Social Networking Solution Based On The TTC Platform</a></li><li><a href="http://www.tigris.finance/">Tigris Protocol - A Secure And Efficient Decentralized Financial Solution Based On the TTC Platform</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7632bb47c98b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>