<?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 Mike Mitchell on Medium]]></title>
        <description><![CDATA[Stories by Mike Mitchell on Medium]]></description>
        <link>https://medium.com/@trigrman?source=rss-3f05355c09f9------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*AayTXGODuJUflbeVHoLwiw.jpeg</url>
            <title>Stories by Mike Mitchell on Medium</title>
            <link>https://medium.com/@trigrman?source=rss-3f05355c09f9------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 11:38:56 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@trigrman/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[Educating the Enterprise on AI]]></title>
            <link>https://medium.com/data-science/educating-the-enterprise-on-ai-f1d206809910?source=rss-3f05355c09f9------2</link>
            <guid isPermaLink="false">https://medium.com/p/f1d206809910</guid>
            <category><![CDATA[education]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[enterprise]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Mike Mitchell]]></dc:creator>
            <pubDate>Thu, 10 May 2018 11:48:28 GMT</pubDate>
            <atom:updated>2018-05-22T19:08:02.142Z</atom:updated>
            <content:encoded><![CDATA[<p><em>The biggest impediment to adoption of AI is lack of knowledge. What do enterprise companies and other ecosystem players need to learn?</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Eopwy3nhWDMpSlFwJbTNYw.jpeg" /></figure><p>It’s easy to say that the promised transformation of every industry by Artificial Intelligence (AI) is exaggerated. The AI buzz extends from an overabundance of business articles (<em>voilá</em>) to the consumer consciousness. Prominent figures in <a href="https://www.vanityfair.com/news/2017/03/elon-musk-billion-dollar-crusade-to-stop-ai-space-x">tech</a> and <a href="http://www.bbc.com/news/technology-30290540">science</a> have gone so far as to express concern about its rapid proliferation.</p><p>Companies dismiss the hype at their peril. Machine learning and other applied AI solutions are indeed having a transformative impact on businesses across industries, and early adopters are seeing more value than those who are just dabbling. A <a href="https://www.mckinsey.com/~/media/mckinsey/industries/advanced%20electronics/our%20insights/how%20artificial%20intelligence%20can%20deliver%20real%20value%20to%20companies/mgi-artificial-intelligence-discussion-paper.ashx">2017 McKinsey Global Institute study</a> found that the more familiar companies become with AI, the more opportunity they see for it in their business. Think new revenue streams, not just cost savings in more departments.</p><p>So, what’s holding many firms back? There are several factors, but two of the biggest impediments to greater adoption are <strong>access</strong> and <strong>knowledge</strong>.</p><p>The access problem is common to technology at this early stage of the adoption curve — clean, comprehensive solutions are still scarce and expensive, and few vendors have a track record. This challenge is now being furiously attacked by mobs of startups and opportunistic carve-outs from established companies. Some are adding value — many are not. You can watch their ebb and flow in various <a href="https://www.techemergence.com/artificial-intelligence-industry-an-overview-by-segment/">segment reviews</a> and <a href="https://appliedai.com/vendors/1">vendor lists</a>.</p><p>But access is not enough. Maximizing value from AI requires a different way of thinking about problems — and about data. To take advantage of the new tools coming available, there are several challenges to be met, and Big Industry cannot overcome these alone.</p><h3>Revolution means rapid change</h3><p>Some will argue we’ve been here before. The Third Industrial Revolution introduced ubiquitous, cheap computing, a major technology-driven upheaval that has barely abated since the 1980s. Industries have adopted and adapted; there have been winners and losers. Why aren’t the current advances just a natural progression instead of something new?</p><p>For one thing, the pace of change in the Digital Age is torrid. Researchers almost universally agree that the world’s data will at least double every two years for the next decade. It’s likely that growth rate will accelerate as storage becomes ever cheaper, and devices of every sort are increasingly connected, generating data to fill it.</p><p>That expansion of data is putting pressure on businesses of every size and industry, creating an urgency to do…something. It turns out that doing something is hard, for a couple reasons.</p><p>First, we have the access problem introduced earlier. Many AI tools available for firms to use in deriving insight from data are still in the research stage. AI applications that can be utilized by a typical business user are targeted at very narrow use cases, and despite that can still be complex and expensive. Increasing flexibility means building a team and developing your own application from scratch.</p><p>But that leads to another problem: training has not caught up to the demand, and the talent pool for AI expertise is shallow. While there’s <a href="https://www.linkedin.com/pulse/really-data-scientist-shortage-scott-nestler/">debate</a> about whether there’s truly a <a href="http://www.itpro.co.uk/careers/28929/data-scientist-jobs-where-does-the-big-data-talent-gap-lie-1">shortage of data scientists</a> today, that debate often doesn’t extend to the need for product managers, operations teams and business strategists who understand how and when AI should be used to their advantage. This entire ecosystem of business functions who are AI-aware is key to its successful application, and that collective awareness is hard to come by.</p><p>Even if a company is able to assemble its own cross-functional AI team, hanging onto them can be a struggle. Especially in organizations with nascent digital capabilities, the culture and process transformation necessary to embrace AI can be slow in coming. Perhaps the data is missing — or the necessary data discipline. Sometimes organizations think they need AI, but really just need data analytics. Or AI might be viewed as a threat to others’ jobs, adding a difficult new dimension to office politics. These situations can result in disillusionment or frustration, and ultimately attrition, for AI teams.</p><h3>Industry needs to go back to school</h3><p>Giving companies access to the latest tools and technology is not a silver bullet for organizations hoping to leverage AI. Instead, we need to be giving people the information they need to make the right decisions about where AI can help them, and how to think critically in this new business context. <em>Knowledge is the key.</em></p><p>Large companies cannot address this challenge alone, though. There are several different groups that need to work together to raise the level of AI awareness in the business ecosystem, and each requires a different kind of knowledge. Let’s look at some questions each of these groups should be asking themselves as they work to build their understanding of concepts that will make them successful.</p><h4>The enterprise needs to rethink their business model</h4><p>While they can innovate on their own, at the top of the food chain the largest organizations are most often consumers of new technology. They benefit most from understanding the big picture, plus the essential details that will drive decision making and direction for their industry or department function. So there are a few key questions that enterprise teams need to be able to answer:</p><ul><li>What are the types of AI, and what are they best used for?</li><li>What data do we need to feed each kind of AI technology, and where do we get it?</li><li>How do we prepare our own data for use by AI?</li><li>Should our focus be on cost reduction, or does AI present a new revenue opportunity?</li><li>Should we hire an in-house data science team, or can our needs be met by a third party black box solution?</li></ul><p>That last question is not quite the same as the traditional “build vs. buy” dilemma. Instead, it’s more about where the knowledge should live. Do you rely on a third party to tell you how to maximize the value of your enterprise data, or should that information always originate within the organization? That brings up one last question, which firms should start thinking about today:</p><ul><li>What’s our data worth, anyway?</li></ul><p>Firms who are able to answer these questions will be the winners of this current industrial revolution. Getting to the answers requires big businesses to internally create and sustain broad, cross-functional understanding of these topics, their specific implications for their industry and impact on customers.</p><h4>Startups and tech companies need to rethink product strategy</h4><p>Startups and smaller, more nimble tech companies are often the innovators and suppliers of new technology to the ecosystem. While they are often better equipped (culturally and personnel-wise, if not financially) for the kind of R&amp;D necessary to come up to speed on new technology, they have their own questions to answer. When thinking about new product offerings, for example:</p><ul><li>How do we create a unique value proposition by adding AI to what we produce today?</li><li>What unique data or algorithms do we have that can drive AI — whether the AI is ours or someone else’s?</li><li>What unique insight do we have on an industry or a process that could be disrupted by AI?</li><li>How do product managers incorporate data and AI into product strategy and the design process?</li></ul><p>That last question hints at a necessary process transformation that product companies will need to undergo. As noted above, there are many startup AI companies emerging to deliver solutions to industry. But we’re still very much in the Wild West era of AI product development, and there’s not yet a standard way for product managers to incorporate data and AI thinking into their designs.</p><p>Product managers already work at the intersection of marketing, design and technology, and the role requires strategic as well as deep analytical thinking, considerable communication capability, and plenty of tact. Is it too much to ask them to also take on data modeling and machine learning algorithms?</p><p>To avoid dumping too much additional responsibility on the product manager role, we will need to form organization structures and processes that incorporate the new design needs generated by the rapid expansion of AI. While they already have a lot on their plates, product managers will need to educate themselves on these topics, and be in the vanguard of the transformation effort.</p><h4>Students and recent grads need to maintain a broad view</h4><p>The third group of ecosystem contributors are the people just entering the workforce. In spite of all the PR earned by companies like IBM and Google about their AI projects, much of the innovation in AI is still happening in academic institutions. So, graduates of data science programs are close to where the magic happens, and are in high demand. But they’re not done with their education just yet.</p><p>As these people begin their business careers, there are a few questions they should be asking themselves, also, such as:</p><ul><li>How do I learn about the diverse variety of possible AI solution sets, and avoid getting locked into a closed ecosystem?</li></ul><p>Expanding knowledge outside favorite technologies and vendors can be difficult whether in academia or business. Big AI platforms from vendors like Microsoft and Amazon are attractive as one-stop shops, but they can be expensive, may not always provide the best data or technology, and vendor lock-in often prevents users from looking for solutions elsewhere. It is to your advantage when prospecting for employment to not only remain neutral, but have broad experience across a range of technology platforms.</p><ul><li>What are examples of how AI has been applied to different problems, and what data was used?</li></ul><p>An understanding of how AI has already been applied in industry solutions will also be extremely useful, in conjunction with recent data science training. As discussed above, businesses are still struggling to understand not only the basic technology, but also how to apply it. So a combination of both theoretical and practical knowledge will be a hot commodity. Furthermore, coming equipped to answer the following question will make you welcome, indeed:</p><ul><li>What can I do to help my colleagues understand where and how to use AI?</li></ul><p>Individuals who have recently spent dedicated time exploring and learning about AI might be best positioned to educate their new colleagues on its uses and benefits.</p><h3>Democratizing AI requires both access and knowledge</h3><p>We’ve explored the rapid expansion of AI technology and its impact, the challenges it raises, and how different parts of the business ecosystem need to educate themselves to address these challenges. While access to technology is growing by leaps and bounds, education is necessary in order to maximize the value of that technology for industry.</p><p>So, industry now needs to establish the institutional knowledge on which the next wave of applied innovation can be built. Without inherent awareness of what kinds of problems AI can solve, what kinds of algorithms can best solve them, and what data is needed to feed those models, the business world will continue to struggle towards the promised automated utopia of the future.</p><p>There’s no quick solution, and enterprise consumers of technology cannot solve the problem alone. Instead there’s a need for a <em>sustained, systemic</em> education effort regarding AI and its applications, building knowledge both in the different segments of current industry, as well as in the emerging workforce.</p><p>Given the threat of a few <a href="https://medium.com/@georgezarkadakis/we-need-to-democratize-ai-and-data-now-c760768f0084">large platforms monopolizing AI technology and data</a>, it’s also important that our new understanding not be skewed to a particular vendor’s perspective, but instead represent as broad a view as possible.</p><p>Everyone should have an equal opportunity to innovate with AI and impact the world for the better.</p><p><em>Companies like </em><a href="https://www.agorai.ai/"><em>Agorai</em></a><em> are working to expand access to AI technology by making best-of-breed tools and data available affordably, not only for the enterprise, but also small and medium-sized businesses. Agorai will also be providing discounted access to its marketplace for educational institutions, and for startups through accelerator programs.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f1d206809910" width="1" height="1" alt=""><hr><p><a href="https://medium.com/data-science/educating-the-enterprise-on-ai-f1d206809910">Educating the Enterprise on AI</a> was originally published in <a href="https://medium.com/data-science">TDS Archive</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>