<?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 Luis Santos on Medium]]></title>
        <description><![CDATA[Stories by Luis Santos on Medium]]></description>
        <link>https://medium.com/@luisfdsantos?source=rss-3d7e7b37e7d7------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*BK5BmPEwL0g7diZGeVbZGg.jpeg</url>
            <title>Stories by Luis Santos on Medium</title>
            <link>https://medium.com/@luisfdsantos?source=rss-3d7e7b37e7d7------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 21:58:26 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@luisfdsantos/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[Revolutionizing Business Decisions: The Power of AI in Forecasting]]></title>
            <link>https://medium.com/@luisfdsantos/revolutionizing-business-decisions-the-power-of-ai-in-forecasting-c2987f8771b3?source=rss-3d7e7b37e7d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/c2987f8771b3</guid>
            <category><![CDATA[forecasting]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[retail]]></category>
            <dc:creator><![CDATA[Luis Santos]]></dc:creator>
            <pubDate>Tue, 30 Jan 2024 08:23:22 GMT</pubDate>
            <atom:updated>2024-01-30T08:23:22.125Z</atom:updated>
            <content:encoded><![CDATA[<p><em>This article is part of a series regarding different AI applications and use cases in business domains.</em></p><p>The concept of forecasting, in its most basic form, has always involved the idea of looking ahead or planning for the future based on current knowledge or expectations.</p><p>Do you remember how sailors in the past used to read the stars and the behavior of the ocean to navigate? Or how traders in ancient marketplaces used to rely on hunches and word-of-mouth to predict market trends?</p><p>Today, forecasting has expanded far beyond that; it now includes forecasting stock markets, election outcomes, disease outbreaks, future sales, product trends, and many others.</p><p>It typically involves using historical data, statistical models, and analysis to make informed predictions about future events or trends.</p><p>The introduction of advanced technologies like artificial intelligence and machine learning has made forecasting more sophisticated, enabling more accurate and detailed predictions across various fields.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k26xV4Pxje2ArNPSa0wNFA.png" /><figcaption>Image generated by the author using DALL-E</figcaption></figure><h4><strong>Some</strong> <strong>Use Cases Across Industries</strong></h4><p><strong>Retail</strong>: forecasting sales, managing inventory, and understanding customer trends.</p><p><strong>Finance</strong>: predicting stock market trends, credit risk assessment, and fraud detection.</p><p><strong>Supply Chain</strong>: anticipating demand, optimizing logistics, and inventory control.</p><p><strong>Healthcare</strong>: predicting disease outbreaks, patient admissions, and resource allocation.</p><p><strong>Energy</strong>: forecasting energy demand and production, especially in renewable energy sectors, to optimize energy generation and distribution.</p><p><strong>Consider the case of Walmart’s AI-powered inventory in retail</strong>. By combining decades of customer service experience and holiday inventory management with AI insights, Walmart is offering a highly satisfying holiday shopping experience for its customers.</p><blockquote>Our AI-powered inventory management system is essential for supplying customers with what they need, when they need it, and at the low costs they expect from Walmart. By leveraging historical data and pairing it with predictive analytics, we’re able to strategically place holiday items across distribution and fulfillment centers, and stores, optimizing the entire shopping experience — <a href="https://tech.walmart.com/content/walmart-global-tech/en_us/news/articles/walmarts-ai-powered-inventory-system-brightens-the-holidays.html">Walmart Global Tech</a></blockquote><p>This approach optimizes stock levels, minimizes waste, and enhances profitability. The accurate forecasting of product demand facilitates efficient inventory management, reducing both excessive stock and shortages.</p><p>By analyzing historical sales data, AI models can identify intricate patterns and trends, uncovering valuable insights into customer behavior.</p><p>These AI systems take seasonal fluctuations into account, keeping an eye on events like holidays, sale promotions, and the launch of seasonal products.</p><p>For example, they can predict the surge in demand for umbrellas during rainy seasons in specific regions or the increased interest in fans during hot summers elsewhere.</p><h4>Strategies for Implementing AI Forecasting in Your Business</h4><p>Given that AI forecasting is an emerging field, it faces several challenges that are yet to be overcome. However, companies that successfully implement AI forecasting can benefit from more precise and dependable predictions.</p><p>Below are some strategies for integrating AI forecasting into your business processes:</p><p><strong>Building a Data-Driven Culture</strong></p><p>To kick things off, companies need to really get into data and make sure their leaders are all about pushing for it. It’s all about working together with data, being open about it, and making sure everyone gets to the data they need to do their job well and get them talking about the best ways to use it.</p><p>A data-driven culture promotes data experimentation and innovation, so encourage your team to play around with different ways of using data to see what works best for the company.</p><p><strong>Bringing in the Experts</strong></p><p>Getting the right pros on board is key when you’re dealing with AI. Given the complexity of this field, it is imperative to engage individuals who possess a deep understanding of AI and can seamlessly integrate it into your existing operations.</p><p>If you’ve got these smart people in-house, great! But if AI feels like a whole new world, don’t hesitate to reach out for external help. It might seem like a big step, but in the long run, it’s worth it. The presence of experienced experts can have a profound impact on the effectiveness of AI implementation within your business.</p><p><strong>Understand your Needs and Define your Goals</strong></p><p>Before diving into using AI for forecasting, it’s important to understand your needs and set your targets. Once you have a clear idea of your objectives, you can start looking into how AI can help you reach them.</p><p><strong>Data Gathering and Preparation</strong></p><p>One of the key aspects of AI forecasting is data collection and harmonization. For algorithms to produce reliable results, you need top-notch data. It’s essential to gather enough historical data so the AI can accurately identify patterns and trends. Whenever you can, enrich your data with external sources like weather forecasts or economic indicators.</p><p>Experts usually face great challenges in working around data limitations and usually perform feature engineering, which is a crucial step.</p><p><strong>Selecting the Optimal Algorithm for Your Needs</strong></p><p>There are many different techniques available for forecasting; it’s important to pick the right one that matches your data and objectives, as each method comes with its own set of pros and cons.</p><p>Different machine learning models are used for predictions, including time-series analysis, regression models, neural networks, and deep learning. Which model to use depends on the type of data you have and your specific forecasting goals, and experts must be the ones deciding on that.</p><p><strong>Implement and Measure</strong></p><p>Once you’ve chosen an algorithm, integrate it into your forecasting framework. Following implementation, monitor the results closely to ensure the AI is delivering accurate predictions, and adjust as needed to fully leverage the power of this advanced technology.</p><blockquote>An excellent case study of implementing a demand forecasting model can be found here: <a href="https://pub.towardsai.net/a-long-term-demand-forecasting-model-implementation-case-study-with-a-major-retailer-35d08342fac0">A Long-Term Demand Forecasting Model Implementation Case Study with a Major Retailer.</a> Despite facing data challenges, this project resulted in a significant 19% decrease in global stock surpluses and shortages for Auchan Retail.</blockquote><h4>Challenges and Considerations</h4><p><strong>Employee Resistance to AI Integration</strong></p><p>Address resistance openly, as the most effective approach often involves engaging in dialogue with people with a negative attitude and trying to understand their feelings and where the apprehension is coming from.</p><p>There are a few things leaders can do: demonstrate the benefits of AI, motivate the team to experiment with it, and affirm that AI is not meant to completely replace human roles.</p><p><strong>Privacy Concerns and Transparency</strong></p><p>I am sure everyone has commented to their friends about how, after chatting about a potential product or searching for it online, their social media is flooded with related ads.</p><p>Some AI tools aim to predict what we will buy a year from now based on our browsing data today. It’s super powerful for businesses, but kind of scary for people. It’s like our computer suddenly knows us better than ourselves, and people do not tend to like it.</p><p>For businesses, it’s crucial to make data collection practices completely transparent to users and clearly explain how algorithms use data to make decisions.</p><p><strong>Data Quality, Quantity, and Bias</strong></p><p>The accuracy of AI forecasting depends heavily on the quality and amount of data available. Moreover, an AI model is only as unbiased as the data it’s trained on, so it’s crucial to train your algorithms on data that does not favor a particular group of people; otherwise, the outcomes will reflect that too.</p><p><strong>Skill Gap</strong></p><p>There is a shortage of skilled professionals capable of comprehending AI and seamlessly integrating it into business strategies, which is expected to persist and even worsen in the coming years.</p><p><strong>Integration with Existing Systems</strong></p><p>Seamlessly integrating AI forecasting tools with existing business systems can be challenging.</p><p><strong>In conclusion, AI-driven forecasting is transforming how businesses plan and make decisions.</strong></p><p>As data sets grow in complexity and size, traditional analysis methods struggle to keep up. That&#39;s why AI’s ability to handle complex, large-scale data and provide precise predictions is invaluable.</p><p>Investing in AI is becoming essential for organizations looking to maintain a competitive edge in the future; however, businesses need to approach this technology with a clear understanding of its capabilities, limitations, and the need for continuous learning and adaptation.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c2987f8771b3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The AI Impact: 8 Key Societal Challenges to Watch in 2024]]></title>
            <link>https://medium.com/@luisfdsantos/the-ai-impact-9-key-societal-challenges-to-watch-in-2024-2cdf328611f5?source=rss-3d7e7b37e7d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/2cdf328611f5</guid>
            <category><![CDATA[deepfakes]]></category>
            <category><![CDATA[society]]></category>
            <category><![CDATA[ethical-ai]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[data-privacy]]></category>
            <dc:creator><![CDATA[Luis Santos]]></dc:creator>
            <pubDate>Mon, 15 Jan 2024 00:25:25 GMT</pubDate>
            <atom:updated>2024-01-15T10:57:50.561Z</atom:updated>
            <content:encoded><![CDATA[<p>Artificial Intelligence (AI) has been around in different types for several years. Yet, like <a href="https://www.youtube.com/watch?v=X-TBhEwnBik">Micah Richards for Man City</a>, 2023 is the year it burst onto the scene thanks to ChatGPT (sorry for the football punt).</p><p>ChatGPT set the record for the fastest-growing user base, <strong>notching</strong> <strong>an estimated 100 million monthly users in just two months</strong>. Facebook, for example, took around four and a half years to hit 100 million users after its launch in 2004.</p><p>The public, myself included, suddenly had the incredible magic of using AI at our fingertips. And with that, the boom was launched 🚀</p><blockquote>Artificial Intelligence will have a greater impact on humanity than fire, electricity, and the internet. — Sundar Pichai, the CEO of Alphabet and Google</blockquote><p>Andrew Ng, one of the most prominent figures in the field of artificial intelligence, compared learning AI to learning literacy in reading and writing, and Elon Musk and top AI researchers wrote an open letter to pause the development of large-scale AI systems, citing fears over the “profound risks to society and humanity”.</p><p>When highly intelligent and influential people highlight the profound impact of AI, it’s only logical to take their concerns seriously and increase awareness about the potential challenges we are likely to face.</p><p>Despite all of this, the technology industry is addicted to boom and optimism and has always dreamt of infinite possibilities and future economic gains.</p><p>Probably 2024 will be a year where companies will find their initial pilots less impactful than they hoped, and that there’s still a lot of hard work to do, and additional funds to invest before dreams become a reality.</p><p>If this growth slows down, it will likely be a refreshing pause, allowing governments, companies, regulators, and hopefully, citizens worldwide to understand AI’s societal impact deeply. If it doesn’t slow down, then we all have a lot to catch up on as quickly as possible.</p><p>It will require much hard work and global collaboration, but it must be addressed immediately.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XD4NY_j3flzcYxNdGPbvaA.png" /><figcaption>AI Wave (image generated by the author with DALL-E)</figcaption></figure><p>Here’s a look at some challenges that we should pay close attention to as we move into 2024:</p><ol><li><strong>Ethical and Privacy Concerns</strong></li></ol><p>When we talk about AI, there are some big questions about how it keeps our data safe and uses information the right way — data privacy and ethical use of data.</p><p>Ethical AI Frameworks are being created, ensuring principles like fairness, accountability, transparency, and respect for user privacy, kind of like a handbook that says AI should be fair, responsible, clear, and careful with our personal info.</p><p>We are witnessing the development of these frameworks by major tech companies — Google, through its Google Cloud service, provides tools like <a href="https://cloud.google.com/explainable-ai">Explainable AI</a>. Similarly, Microsoft is contributing with its <a href="https://blogs.microsoft.com/wp-content/uploads/prod/sites/5/2022/06/Microsoft-Responsible-AI-Standard-v2-General-Requirements-3.pdf">Microsoft Responsible AI Standard</a>. Startups like <a href="https://www.fiddler.ai/">Fiddler</a> and <a href="https://www.profian.com/">Profian</a> are also focusing on making AI transparent and trustworthy.</p><p>Companies that follow an ethical use of AI are increasingly adopting techniques that protect privacy and personal information in AI datasets:</p><ul><li><strong>Advanced data anonymization and encryption</strong> — ensuring that AI models are trained on data that has been anonymized, especially when dealing with sensitive personal information.</li><li><strong>Data minimization principles </strong>— collecting only the data necessary for the specific purpose for which it is processed.</li></ul><p>Daniela Braga, CEO of <a href="http://defined.ai/">Defined.ai</a> — whom I am a huge fan of, not just as a proud Portuguese thrilled to follow a Portuguese woman, member of a 12-person body that advises the US president on the AI strategy of the United States, but also because I completely relate to her view of ethical AI — said this at the start of this year:</p><blockquote>Starting in 2024, let’s remind ourselves that every AI begins with data.</blockquote><blockquote>The dataset, not the model’s architecture or hyperparameters, fundamentally shapes an AI’s behavior. This is why we have an immense responsibility in dataset construction.</blockquote><blockquote>Many companies rely solely on open-source datasets, which don’t have commercial licenses to build AI products. While they are valuable resources, they often lack the depth, diversity, and proper license required for foundational models’ deployment.</blockquote><blockquote>At <a href="https://www.linkedin.com/company/definedai/">Defined.ai</a>, we’re committed to elevating the standard for dataset creation. They must undergo rigorous auditing to ensure their integrity. Consent and fair compensation for contributors are non-negotiable, reflecting our dedication to ethical data practices. We’re proud of our diverse and inclusive approach, involving a wide range of participants in dataset creation.</blockquote><p>This should be the standard moving forward.</p><p>Raising public awareness about data rights and the implications of AI is critically important, and I am hopeful that significant efforts will be directed towards this in 2024.</p><p>2.<strong> AI Bias and Discrimination</strong></p><p>AI systems can perpetuate or amplify biases if they’re not designed carefully — this could lead to unfair practices or discrimination.</p><blockquote>Some AI-driven recruitment tools have shown biases against women. For instance, an AI system used by Amazon in 2018 for screening job applicants reportedly favored male candidates over female candidates, particularly for technical jobs, due to biases in the training data which was based on resumes submitted over 10 years, mostly by men.</blockquote><p>Similarly, numerous studies and reports have also shown that certain facial recognition systems are less accurate in identifying individuals with darker skin tones.</p><p>This has raised significant concerns, especially regarding the use of these technologies in law enforcement and legal settings, potentially leading to wrongful arrests or misidentification, amongst other major upsets.</p><p>These are just a few examples. Avoiding bias in AI systems is a multifaceted challenge that requires attention at various stages of AI development and deployment. Some initiatives should include:</p><p><strong>Diverse and Representative Data</strong> — ensure that training data is diverse and representative of different groups. We already see some examples from leading companies:</p><ul><li><strong>IBM launched the Diversity in Faces (DiF) project</strong> to create a dataset that helps build more diverse and accurate facial recognition systems. The DiF dataset is designed to be diverse across multiple dimensions, such as age, gender, skin tone, and facial features.</li><li><strong>Google organized the Inclusive Images Competition</strong> to encourage the development of more inclusive image classification models. The competition challenged participants to create models that perform well across a diverse set of images, representing different geographic regions and demographics.</li></ul><p>Other crucial topics, including having diverse development teams, educational training on understanding and mitigating biases, regular third-party bias audits, and public accountability, need to be discussed more extensively and will gain increasing relevance in the coming years.</p><p>3. <strong>Deepfake and Misinformation</strong></p><p>In this digital age, the emergence of deepfakes and rampant misinformation poses unprecedented challenges.</p><p>Deepfakes are AI-generated fake videos or audio recordings that appear shockingly real and, while technologically impressive, have facilitated a new era of misinformation.</p><p>According to a recent <a href="https://www.weforum.org/agenda/2024/01/global-risk-report-2024-risks-are-growing-but-theres-hope/">report from the World Economic Forum</a>, the spread of false information powered by AI technologies is <strong>the most immediate risk to the world (!) in 2024</strong>.</p><p>Some examples of deepfakes risks:</p><p><strong>Threats to Democracy</strong> — with over 40 significant elections on the horizon in 2024 (<a href="https://english.elpais.com/international/2023-12-19/2024-when-half-of-the-planet-goes-to-the-polls.html#">around half of the world’s adult population heads to the polls!</a>), there’s growing concern about the potential misuse of new technologies by political figures, activists, and even external entities aiming to shake election outcomes.</p><p>By manipulating public perception, deepfakes pose a direct threat to democratic processes, and can be used to catch voter’s attention, as highlighted in Argentina:</p><blockquote>In the final weeks of campaigning, Argentine President-elect Javier Milei published a fabricated image depicting his Peronist rival Sergio Massa as an old-fashioned communist in military garb, his hand raised aloft in salute. The apparently AI-generated image drew some 3 million views when Milei posted it on a social media account, highlighting how the rival campaign teams used artificial intelligence technology — <a href="https://www.context.news/ai/how-ai-shaped-mileis-path-to-argentina-presidency"><em>Context</em></a></blockquote><p><strong>Psychological impact and reputation damage on victims</strong> — for individuals who become targets of deepfakes, the psychological impact can be profound, as victims often experience emotional distress, damage to their reputation, and in some cases, threats to their physical safety.</p><p>I recently heard about a distressing incident in Brazil from a friend. In this situation, young kids aged sixteen used a deepfake app that undresses women on a female colleague (the so-called <em>nudify apps</em> that are soaring in popularity) and took a photo that was widely shared via WhatsApp among students at a school.</p><p>Despite later clarifications that the image was not real, it had a deeply negative psychological and reputational effect on the poor girl, with concerns about lasting impact in the future.</p><p>These are some examples but many more can be created by deepfakes, such as online scams, frauds, blackmailing, and violence incitation.</p><p>As expected, AI-driven tools are being created to detect and combat deepfakes.</p><blockquote>McAfee announced an AI-powered Deepfake Audio Detection tool. ‘Project Mockingbird’ reportedly achieved a 90% success rate in identifying fraudulent audio to protect users from sophisticated online scams.</blockquote><p><a href="https://realitydefender.com/">Reality Defender</a> — an advanced AI-powered platform designed to identify and combat deepfakes — offers a comprehensive solution to detect and manage AI-generated manipulations in various media formats, including audio, video, images, and text.</p><p>Blockchain technology is also emerging as a method to verify the authenticity of digital media, creating an immutable record of the original content.</p><p>At the same time, social media platforms and content distributors play a crucial role in monitoring and removing deepfake content from their services.</p><blockquote>Meta Platforms, which owns Facebook and Instagram, said that from 2024 advertisers will have to disclose when <a href="https://www.reuters.com/technology/meta-require-disclosures-ai-created-altered-political-ads-2023-11-08/">AI or other digital methods are used</a> to alter or create political, social or election related advertisements on the sites.</blockquote><p>Meta Platforms also began blocking keywords associated with searching for undressing apps, trying to solve the problem mentioned above.</p><p>Finally, educating the public about the existence and nature of deepfakes is essential in the fight against misinformation.</p><p>Through media literacy, campaigns can teach people to critically evaluate the content they consume and share online. How many times are we scrolling on social networks and see information we know is not true or framed out of context shared to get attention? With deepfakes they will happen more often.</p><p>Awareness programs can guide individuals on how to verify information, understand the source’s credibility, and be cautious about sharing unverified content.</p><p>Some folks, myself included, think that the emergence of unreliable AI-created content might boost journalism’s and influencers&#39; credibility and trust. Others, however, fear it could lead to a general distrust in all information.</p><p>2024 is expected to provide more insights into which of these outcomes might occur.</p><p>4. <strong>Legal and Regulatory Challenges</strong></p><p>As highlighted above, adapting legal frameworks to keep up with AI advancements such as deepfakes poses a significant challenge.</p><p>AI advances rapidly, but as we know, governments and public institutions often tend to be slow and bogged down in bureaucracy.</p><p>However, regulators should revise current legal frameworks to include considerations for AI and its impacts and develop and implement regulations specifically designed for AI technologies.</p><p>We should also encourage an interdisciplinary approach to AI law and regulation, combining insights from legal experts, technologists, ethicists, social scientists, amongst other specialists.</p><p>Global cooperation will be key — if governments across the world agree on how to use AI and what’s ethical, we can have laws that work well everywhere. It will probably not be that simple huh?</p><p>Yet, the urgency of this matter cannot be understated.</p><p>When things go wrong, who is responsible? The developers? The users that consent? Or the AI system itself?</p><p>Let’s take a look at some examples:</p><ul><li>If an autonomous vehicle is involved in an accident, it can be difficult to determine whether the fault lies with the AI system’s decision-making, a malfunction in the vehicle’s sensors or hardware, the actions of other drivers, or even the road conditions.</li><li>When AI algorithms are used to moderate content on social media platforms and end up wrongfully flagging or failing to flag inappropriate content, determining responsibility can be complicated. It could be due to flaws in the AI’s understanding of context, the guidelines set for moderation, or even the quality of the data used to train the AI.</li><li>Another example related to AI systems used to recommend treatments based on patient data. If a treatment proves ineffective or harmful, it could be challenging to determine if the cause was an error in the AI’s analysis, inaccurate or incomplete patient data, or if the AI was not updated with the latest medical research.</li></ul><p>These examples underscore the significant challenges we face in regulating such complex systems.</p><p>As complex as it is, some initiatives have already taken place in the European Union (EU) and the United States (US):</p><p>In the EU, the AI Act intends to be the world’s first comprehensive AI law to ensure that AI systems are <strong>safe, transparent, traceable, non-discriminatory, and environmentally friendly</strong>.</p><p>This legislation focuses on classifying AI systems based on the risk they pose, categorizing them into different risk levels:</p><ul><li><strong>Unacceptable risks </strong>— such as those that manipulate behavior or engage in social scoring, will be banned.</li><li><strong>High-risk AI systems </strong>— like those used in critical infrastructure or law enforcement, will be subject to strict assessments before and during use.</li><li><strong>Limited risk</strong> — such as a Generative AI, also includes transparency requirements and provisions.</li></ul><p>In the US, President Joe Biden has also initiated efforts to address AI advancements, focusing on creating guidelines and policies that ensure responsible development and use of AI.</p><p>Recently, t<a href="https://www.iso.org/home.html">he ISO — International Organization for Standardization</a> has introduced the first AI management system standard, ISO/IEC 42001.</p><p>With 127 countries contributing, it represents a global effort, and companies like Amazon Web Services (AWS) are already aligning with this new standard.</p><p>It’s something to follow in 2024.</p><p>5. <strong>Skill Gap and Reskilling Challenges</strong></p><p>Very related to my previous article ‘<a href="https://medium.com/@luisfdsantos/ai-the-catalyst-for-a-new-era-in-the-workplace-1c6a0a8f510b">AI: The Catalyst for a New Era in the Workplace</a>’, where I explored the rapid evolution of AI and its impact on the job market.</p><p>This skill gap has been already a reality over the last few years with the lack of people prepared to work in the technology field but will be aggravated with AI.</p><p>I have been working in reskilling and upskilling for the last few years, and have witnessed the positive impact on people, businesses, and societies in general.</p><p>In one of the reskilling projects I have been heavily involved with <a href="https://codeforall.com/">Code For All_</a>, we had the goal to help build a tech hub from scratch in the middle of the Atlantic Ocean, on a small Azorean island called <a href="https://maps.app.goo.gl/9TwdyaSZP8XA2EcSA">Terceira</a>.</p><p>During the Trump administration, the Pentagon withdrew a large part of the North American contingent from Lajes Base, impacting the socio-economic context of this small island.</p><p>The goal was to mitigate the social and economic effects of the downsizing, creating qualified employment through the creation of a technological hub for companies dedicated to software development — from the Azores to the world.</p><p>Over the four years of the project, the results were promising: we attracted over 20 companies and generated more than 200 qualified jobs, thereby making a direct impact on the socio-economic landscape.</p><p>These results exemplify the positive impact of public-private partnerships in more remote and unfavoured areas. While many success stories exist, there’s still a significant need for more.</p><p>We should encourage partnerships between educational institutions, governments, and businesses to ensure skill training that aligns with current market demands, extending beyond the tech field to other areas with identified skill gaps.</p><p>6.<strong> Future of Education</strong></p><p>With the rise of AI, the education system may need to evolve, focusing more on developing technical skills and other skills that AI cannot replicate easily, such as creativity, critical thinking, and emotional intelligence.</p><p>In many public institutions, certain fields are perceived as being behind the times, lacking adaptation to modern technological advancements, and not updated in terms of skills and knowledge, especially in areas crucial for the future workforce.</p><p>Particularly, the deficit in essential skills, like understanding the current and emerging technologies, programming languages, and digital tools is clear.</p><p>In today’s digitally driven world, these skills are no longer just advantageous but necessary. It’s about preparing individuals not just for the jobs of today but for the evolving challenges and opportunities of tomorrow.</p><p>Simultaneously, it’s essential to invest in training educators, equipping them with the necessary skills to effectively teach these subjects.</p><p>Integrating AI tools into their teaching methodologies can further enhance the learning experience and ensure that education keeps pace with technological advancements</p><blockquote><a href="https://www.khanacademy.org/">Khan Academy</a> has introduced Khanmigo, an innovative AI-powered tool designed to enhance the learning experience. Khanmigo serves as a personalized tutor for learners, offering one-on-one support, fostering critical thinking, and suggesting relevant educational resources.</blockquote><p>Also for teachers, Khanmigo serves as an assistant for various tasks — it helps in making lesson plans, provides insights into how students are doing, and assists in creating teaching goals and assessments.</p><p>It’s also essential to shift our focus to promoting lifelong learning and ensuring continuous education opportunities. This approach will empower individuals to keep up with rapid technological advancements and adapt to the ever-evolving job market.</p><p>The transformative power of AI in education is vast. Particularly, it has the potential to reduce barriers to education, making learning more accessible to students regardless of their socioeconomic status, location, or other factors.</p><p>Prestigious organizations like <a href="https://hai.stanford.edu/events/aieducation-summit-ai-service-teaching-and-learning">Stanford HAI</a> and <a href="https://www.unesco.org/en/digital-education/artificial-intelligence">UNESCO</a> have been providing a wealth of information on the current state and future possibilities of AI in education, highlighting its potential benefits and the challenges that need to be addressed.</p><p>As we move into 2024 and beyond, the need for such informed, multi-dimensional perspectives will become increasingly important.</p><p>7.<strong> Human Dependency on Technology</strong></p><p>The prospect of becoming overly reliant on AI systems and technology is a real concern, as it could lead to a diminution of critical human skills and judgment.</p><p>I am a big fan of AI, and my enthusiasm only grows as I consider its vast possibilities. However, I am a fan of AI as a tool to augment human abilities, not replace them.</p><p>Of course, this can become harder as we accept and adopt AI in our lives, but because we are still in the early days, we should encourage the development and use of AI systems that require human oversight and input, cultivating a collaborative relationship between humans and technology.</p><p>Similarly, businesses should critically evaluate, before implementing AI solutions, whether the solution enhances human work or unnecessarily replaces it. Of course, money will have a stronger influence, but we should still strive for it.</p><p>As technology takes over a big part of our lives, as we are already experiencing today, we should encourage more human interaction:</p><ul><li><strong>In workplaces</strong>, encourage team collaboration and face-to-face interactions to maintain and develop social skills and team dynamics.</li><li><strong>In educational settings</strong>, emphasize group work and social activities to balance technology use with human interaction.</li></ul><p>We should keep educating the public and our kids, about the importance of limiting AI and technology use in personal life to maintain a healthy balance, like setting boundaries for screen time and the use of digital assistants.</p><p>However, as the image below suggests, this balance is far from being achieved, is it?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NTtFryRGouVI5exTZTJl5g.jpeg" /><figcaption>Everyone is on their phones watching the New Year’s Eve countdown (<a href="http://alamy.com/">source</a>). You can check the video on <a href="https://www.tiktok.com/@dailymail/video/7319609396693077279?lang=en">Daily Mail Tik tok</a>.</figcaption></figure><p>8. <strong>Environmental Impact</strong></p><p>The increased use of AI and technology requires energy and resources, potentially impacting environmental sustainability.</p><p>AI systems, especially large data centers and advanced models like deep learning, consume significant amounts of electricity, contributing to higher energy demands and carbon emissions.</p><p>Carbon Footprint of training complex AI models can emit substantial amounts of CO2, equivalent in some cases to the lifetime emissions of several cars.</p><p>Some examples:</p><ul><li>In 2021, Google DeepMind’s AI system used approximately 652,000 kWh of energy annually, which is almost eight times the average energy consumption of a UK household.</li><li>Similarly, the energy used for the initial training of OpenAI’s GPT-3 Model was equivalent to the amount produced by burning 9 million pounds of coal.</li></ul><p>It will be important to promote research and development in energy-efficient AI algorithms and data centers to reduce this environmental footprint.</p><p>For example, researchers are investigating methods to make AI models less complex while maintaining their effectiveness. They’re using strategies like trimming down the model (model pruning), simplifying the data it uses (quantization), and designing more streamlined neural networks.</p><p>It will also be important to encourage AI and technology companies to power their operations with renewable energy sources such as solar and wind power, as well as converting waste heat back into electricity. Additionally, nuclear energy is often regarded as an optimal source of alternative energy.</p><p>Finally, for this to be a reality governments need to implement regulations and provide incentives for sustainable practices in the tech industry.</p><p>Leading tech companies heavily invested in AI are publicly committed to sustainability. For instance, Microsoft’s goal is to become carbon-negative by 2030 and to offset all of its historical emissions by 2050. Meanwhile, Google has set a target to operate on completely carbon-free energy by 2030.</p><p>Despite some eco-friendly AI initiatives, there’s a need for broader advancements in the AI industry to ensure that both the foundational development and the infrastructure of AI systems are inherently more environmentally friendly.</p><p><strong>Final Thoughts</strong></p><p>Talking about problems is often easier than actually solving them.</p><p>Important areas like education have been neglected for years and haven’t received enough public attention. So, it’s probably not realistic to expect quick action in areas that the public still doesn’t fully understand or know much about like AI.</p><p>However, the influence of AI on society is expected to be huge.</p><blockquote>The tools and technologies we’ve developed are really the first few drops of water in the vast ocean of what AI can do.</blockquote><blockquote><em>Fei-Fei Li, AI Researcher &amp; Professor, Stanford University</em></blockquote><p>I believe significant changes are on the horizon in the coming years, and I’ll continue to write and raise awareness about them.</p><p>By next year, I hope to reflect on this article and observe how these challenges have evolved and what solutions have emerged to address them.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2cdf328611f5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI: The Catalyst for a New Era in the Workplace?]]></title>
            <link>https://medium.com/@luisfdsantos/ai-the-catalyst-for-a-new-era-in-the-workplace-1c6a0a8f510b?source=rss-3d7e7b37e7d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/1c6a0a8f510b</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[careers]]></category>
            <category><![CDATA[work]]></category>
            <category><![CDATA[automation]]></category>
            <dc:creator><![CDATA[Luis Santos]]></dc:creator>
            <pubDate>Tue, 02 Jan 2024 17:21:23 GMT</pubDate>
            <atom:updated>2024-01-02T17:21:23.747Z</atom:updated>
            <content:encoded><![CDATA[<p>My growing passion for Artificial Intelligence (AI), fueled by my experience in assisting over a thousand individuals in their job searches over the last few years, has sparked a deep curiosity about the future of work in the AI era. This article represents my quest to understand and adapt to the changing employment landscape, ensuring I can continue to effectively guide others in their career paths and surfing the AI wave.</p><p><strong>The AI Job Apocalypse: Myth or Reality?</strong></p><p>Recently, I attempted to explain AI and what I have been learning to my mom. Her response was a traditional one: “OMG, we are going to be replaced and dominated by robots like in the movies.”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/512/1*TFm1leunN4gSuWDtsKWN6Q.jpeg" /><figcaption>Image generated by DALL-E</figcaption></figure><p>I shared my view with her that the robots portrayed in movies are unlikely to replace humans in the near future. However, I also showed some examples of the current AI technologies that are already replacing human roles, which surprised her.</p><p>This reaction is common among people who are not heavily involved with technology or lack interest in the subject. It is important to raise awareness about these advancements, as they are already reshaping our world and will continue to have a significant impact on our future.</p><p>Throughout history, new technologies have often sparked more apprehension than optimism, as workers feel threatened by machines.</p><p>In the 19th century, approximately 80% of the workforce was involved in agriculture. This number rapidly decreased to just 2% as a result of mechanization. However, those who were displaced were able to find new employment opportunities in the emerging industry sector.</p><p>In the same era, during the Industrial Revolution, English workers known as Luddites infamously destroyed textile machines out of fear of job displacement. In the end, machines did indeed take over their jobs, but they managed to find employment elsewhere.</p><p>In these examples, the only thing that changed was the type of machine, but the situation remains the same.</p><p>Initially, unemployment was created, but it also led to the creation of <strong>new partnerships between</strong> <strong>humans and machines</strong>. Something similar might be happening now.</p><blockquote><strong>BYOAI (Bring Your Own Artificial Intelligence)</strong> is a growing trend in offices where employees use their own AI tools at work. This is becoming popular because AI tools are now more affordable, easier to use, and more and more jobs need AI skills.</blockquote><p>I am an optimist by nature, and I am an optimist in this case.</p><p>Working with new tools has always improved our lives, and technology has played a crucial role in our evolution. I believe AI will keep enhancing our efficiency as human beings.</p><p>We are creative people by nature and tend to prosper by solving problems with new technologies. As long as we keep our nature, there is no reason to believe it will change this time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/376/1*ji2mlqjaou6ARlf4l0bFmw.jpeg" /><figcaption><em>Fiverr announced a brand campaign called “Power of Humanity.” As the world of work quickly evolves with the rise of AI technology, this campaign recognized and celebrated the essential role humans (still) play.</em></figcaption></figure><p>Optimism aside, problems will arise.</p><p>To date, technology has had a positive impact on the job market. Therefore, one might think that further improvements through AI will enhance this trend. However, with the rapid advancements in AI, it’s uncertain what the future holds.</p><p>Robots are already replacing personal assistants.</p><p>Algorithms are writing articles for newspapers.</p><p>Artists and writers are possibly the next in line.</p><p>New York Times filing a lawsuit against Open AI and Microsoft for copywriting violation, alongside similar actions by other intellectual property holders, underscores the significant impact of AI.</p><p>Even autonomous cars may soon replace taxi drivers (a lot of question marks about when this will be a reality tho).</p><p>Those were some of the examples I explained to my mom, countering the more dramatic, Hollywood-style narrative of robots taking over society.</p><p>Thus, the question arises: Is a work apocalypse approaching?</p><p>In my optimistic view, I don’t see the emergence of AI as stealing jobs. It eliminates repetitive tasks and makes them more efficient.</p><p>However, if those repetitive tasks are the main part of your job, I think you should start worrying now. There are jobs that automation will swallow, and the general trend is clear.</p><blockquote>We can take the example of Microsoft Copilot, a revolutionary AI-powered tool designed to enhance productivity and creativity across various aspects of work and personal life. Seamlessly integrated with Microsoft’s suite of applications, it automates routine tasks, offers intelligent data analysis, and provides insightful assistance in content creation and decision-making processes.</blockquote><p>I might even be underestimating by suggesting that AI will primarily eliminate repetitive tasks, because there’s a risk that AI could displace more complex jobs too, leading to a larger impact on employment than anticipated. But let’s keep that for future chapters.</p><p><strong>Redefining Work in the Era of Automation</strong></p><p>According to the World Economic Forum, employers are anticipating that between 69 to 83 million new jobs will be created and 673 million jobs that currently exist will be eliminated.</p><p>Take the example of Robotic Process Automation (RPA), and how is revolutionizing the workplace by automating routine and time-consuming tasks.</p><p>How can a human compete with a system that can work 24/7, providing unparalleled efficiency and consistency?</p><p>Consider the role of an individual in an accounting department, for instance:</p><blockquote>RPA systems can read invoices, input data into the correct fields, send confirmation emails, and even handle discrepancies without any human intervention while working around the clock.</blockquote><p>The back office can be the ideal place to start with automation, since it is not part of the customer support service domain and is where the company’s main operations are carried out, such as HR and Finance.</p><p>They are very susceptible to automation since a lot of data is collected in them, and many numbers can be transformed into meaningful information.</p><p>Despite displacing some jobs, automation should not be viewed solely as a threat to employment but also as an opportunity to enhance human potential.</p><p>It will let workers focus on more creative, strategic, and meaningful tasks and reduce errors in boring administrative tasks.</p><p>Undoubtedly, there will be a lot of societal challenges such as unemployment and a reorientation of skills, but this will also signify new opportunities for many individuals who have yet to realize their potential.</p><p>Employees or individuals with STEM (Science, Technology, Engineering, Math) skills will be in high demand. So if you’re already on this path keep going and improve other areas. For the others, here are some pieces of advice:</p><blockquote>Even if it seems distant, the reality is some industries and jobs die overnight. To avoid becoming one of these obsolete roles and survive in the job market, you must be adaptable. You must have a good sense of what’s to come and then take the initiative.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KeToUM5Yv4c-9QyCDeRZ7w.png" /><figcaption>Image created by DALL-E</figcaption></figure><p>Despite the intimidating changes, the best way to prepare is to become a well-rounded candidate with strong human skills.</p><p>A study by the OECD based on research from the World Economic Forum shows that emotional intelligence, solving complex problems, critical thinking, and creativity will be among the top ten skills most demanded that employers will look for in 2025.</p><p>Can we all agree that little or effective training has been created for these skills?</p><p>To mitigate the impact of job displacement, there is a critical need for upskilling and reskilling programs. These initiatives should focus on equipping the workforce with the skills required for the jobs of tomorrow.</p><p>In my opinion, is not even a question of whether AI will displace jobs that rely heavily on routine and manual tasks, but when. As we have seen it’s already happening, but it will be much bigger than that.</p><p>If you’re someone working in this type of field, I would strongly suggest you start thinking about your life for the following years.</p><p><strong>What’s Next for Work? Exploring Potential Future Professions</strong></p><p>The rise of AI technologies is expected to create new categories of jobs, many of which will revolve around the development, monitoring, and maintenance of these systems, as well as roles that leverage human capabilities that AI cannot easily replicate.</p><p>There will be a lack of people for those positions and certainly, transitioning to new AI-related roles often will involve leveraging skills from traditional backgrounds and enhancing them with new knowledge.</p><p>For many of these transitions, additional training in AI, machine learning, data analytics, or a related field will be highly beneficial. This could come in the form of online courses, bootcamps, certifications, or even advanced degrees, depending on the role and the individual’s starting point.</p><p>The key is to combine existing domain expertise with new tech skills to stay competitive in an AI-augmented job market.</p><p>Here are some potential jobs that might emerge, with traditional backgrounds and transferable skills that could be an advantage to perform:</p><p><strong>AI Trainer</strong></p><p>Individuals who train AI systems to perform tasks such as recognizing speech or images, understanding natural language, or even emulating human behaviors. This includes teaching AI context, cultural nuances, and ethical operations.</p><ul><li>Traditional Background: Psychology, Education, Communication</li><li>Transferable Skills:<strong> </strong>Understanding of human behavior, communication skills, teaching experience</li></ul><p><strong>AI Maintenance Specialists</strong></p><p>Experts who service and repair AI systems, ensuring they operate efficiently and effectively, much like IT support for software today but with a focus on AI-specific issues.</p><ul><li>Traditional Background: IT Support, Computer Science</li><li>Transferable Skills: Technical troubleshooting, hardware and software knowledge</li></ul><p><strong>AI Ethicists</strong></p><p>As AI becomes more prevalent, there will be a growing need for professionals who can address the ethical implications of AI. This could include formulating guidelines for the ethical use of AI, auditing algorithms for bias, and ensuring that AI systems align with societal values.</p><ul><li>Traditional Background: Philosophy, Law, Social Science</li><li>Transferable Skills: Ethical reasoning, regulatory knowledge, social research</li></ul><p><strong>Human-AI Interaction Designers</strong></p><p>These professionals will design interfaces and experiences that enable smooth interactions between humans and AI systems, from voice-activated assistants to collaborative robots (cobots).</p><ul><li>Traditional Background: UI/UX Design, Cognitive Science, Human-Computer Interaction</li><li>Transferable Skills: Design thinking, user research, empathy for user experience</li></ul><p><strong>Personal Privacy Advisors</strong></p><p>Professionals who help individuals manage and protect their personal data and advise companies on data collection and management practices, in compliance with privacy laws and regulations.</p><ul><li>Traditional Background: Law, Information Technology, Cybersecurity</li><li>Transferable Skills: Knowledge of privacy regulations, security protocols, advocacy skills</li></ul><p><strong>Autonomous Transportation Specialists</strong></p><p>As autonomous vehicles of all kinds become more prevalent, specialists will be needed to oversee fleets, manage logistics, and navigate the legal and practical aspects of these technologies.</p><ul><li>Traditional Background: Urban Planning, Transportation Engineering, Logistics</li><li>Transferable Skills: Knowledge of transportation systems, project management, regulatory awareness</li></ul><p><strong>Digital Therapists</strong></p><p>Professionals who assist individuals in managing the psychological and social effects of an increasingly digital and AI-driven world, helping them find a balance between tech and life.</p><ul><li>Traditional Background: Psychology, Counseling, Wellness Coaching</li><li>Transferable Skills: Empathetic listening, behavioral therapy techniques, stress management</li></ul><p><strong>Algorithm Bias Auditors</strong></p><p>Professionals who scrutinize algorithms for biases that can lead to discrimination and work, to ensure fairness and inclusivity in automated decisions.</p><ul><li>Traditional Background: Data Science, Actuarial Science, Sociology</li><li>Transferable Skills: Data analysis, critical thinking, knowledge of statistical biases</li></ul><p>Those are just some examples.</p><p>Some of them might not happen or happen in different shapes.</p><p>Some others will arise.</p><p>The actual titles and responsibilities of these roles will evolve with the technology and the needs of the market.</p><p>As AI and automation change the employment landscape, the emphasis on lifelong learning and adaptability will be crucial for the workforce to transition into new opportunities.</p><p><strong>Humans are (still) the answer</strong></p><p>Having said all of that, one reason why I think robots are unlikely to replace us soon is that the touch and consideration of humans are increasingly a selling point.</p><p>In my particular case, I don’t like being surrounded by algorithms that know me better than I do, and I am increasingly concerned about the data I provide. I believe that more people think this way and there is a greater awareness of this topic.</p><p>Companies are also aware of it and are working to include it in their strategies.</p><p>Even in the health area, which has been mentioned so much for its advances, humans will continue to be indispensable for their emotional sensitivity.</p><p>Can you imagine a robot caring for you in a moment of distress, when we need affection the most?</p><p>However, we cannot deny that AI is threatening jobs at an increasing speed, and technologies can make entire sectors obsolete.</p><p>As soon as many enterprises start to use it, others will necessarily have to do the same or they will fall behind and cease to be competitive.</p><p>It will be very difficult to predict when this shift will occur, but in the long run, both individuals and societies must consider these scenarios.</p><p>I’ll be following the upcoming chapters closely and sharing them. I encourage everyone to do the same.</p><p><strong><em>Disclaimer:</em></strong><em> As an AI enthusiast on a learning journey, the insights I share here come from my personal perspective. The field of AI is a vast and evolving landscape, filled with uncertainties even for “AI Experts”. If you’re keen on following my learning path and ideas, please follow along.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1c6a0a8f510b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>