Inclusionism

James Felton Keith

James Felton Keith explores the tenets of inclusionism . As always, JFK is joined by Peter Willumsen and friends. His ethic of inclusionism is at the core of his work on global diversity equity inclusion and belonging. JFK is the Founder of Techstars Venture Backed, InclusionScore Companies and a lecturer at the University of Georgia's Terry College of Business. In 2017 he became the first Black LGBT person to run for federal office in the United States during his bid for Congress to represent Harlem. jamesfeltonkeith.substack.com

  1. The White House is Sacrificing People for Cold War AI “Dominance”.

    23 мар.

    The White House is Sacrificing People for Cold War AI “Dominance”.

    The White House has now made its position on artificial intelligence much clearer. It sees the harms. It sees the scams, the deepfakes, the threats to children, the pressure on the electric grid, the fights over copyright, and the disruption coming for workers. It is not blind to the damage. But it is still committed to the same old framework of power: sacrifice the public, protect the market, and call it national strength. That is the deeper problem with this AI policy framework. It talks about the visible harms of AI, but it refuses to confront the extraction model underneath it. It wants to regulate the symptoms while preserving the structure. It wants America to dominate AI, but it never really asks who pays for that dominance and who gets left behind to make it possible. That idea of “dominance” is not neutral. It comes out of a Cold War mindset. It assumes that the nation must race ahead at all costs, that scale itself is victory, and that any serious constraint on industry is a strategic weakness. In that worldview, people are not partners in development. They are fuel. Workers are a labor supply problem. Families are a safety issue. Communities are an infrastructure problem. States are obstacles. The public is something to manage while the machine expands. That is exactly what this framework sounds like. It offers protections at the margins. It talks about guarding children, fighting fraud, and limiting some of the most obvious abuses of synthetic identity. Fine. Those are real problems. But those protections do not change the central fact that AI systems are being built on human expression, human behavior, human culture, and human decision-making without creating meaningful economic rights for the people whose lives are being turned into inputs. That is the scandal. The White House is willing to admit that AI can harm people. It is not willing to admit that AI is profitable because it extracts value from people. So the framework treats the public as something to defend, not something to pay. It treats workers as people who need retraining, not as contributors who deserve bargaining power. It treats communities as places that should be protected from higher electric bills, not as stakeholders who should have a claim on the wealth being generated around them. It treats creators as possible rights-holders, but ordinary people remain economically invisible even though their data, patterns, preferences, and behavior are part of what makes these systems useful. That is not reform. That is damage control. And it is noticeably behind the rest of the world. The European Union has already done what Washington still seems afraid to do: pass an actual AI law. Europe may not have solved everything, but it at least recognized that artificial intelligence is important enough to govern in binding terms. It built a legal structure. It imposed obligations. It drew lines. The United States is still talking like a superpower in decline: obsessed with beating rivals, terrified of slowing industry, and unwilling to discipline capital even when the public cost is obvious. That is what makes this framework feel so dated. Its language is modern, but its politics are old. Beneath the talk of safety and innovation is a very familiar national logic: centralize power, protect incumbents, preempt local resistance, and tell the public that sacrifice is necessary for the greater good. We have heard that story before. We heard it in the name of industrial supremacy. We heard it in the name of military necessity. We heard it in the name of global competition. And ordinary people were almost always told to wait, adapt, trust the experts, and accept the tradeoffs. Now they are being told the same thing again in the language of AI. The public should not accept it. If artificial intelligence is built from society, then society must have rights in what it produces. Not just warnings. Not just safety promises. Not just technical standards written by the same firms racing to dominate the field. Rights. Compensation. Bargaining power. Ownership. Anything less leaves the basic arrangement untouched: the people generate the value, the companies capture the upside, and the government manages the fallout. That is why this framework is not bold. It is cautious where it should be transformative. It is aggressive where it should be democratic. It is protective only at the edges, while the center of the system remains deeply extractive. America does not need a Cold War theory of AI dominance. It needs an AI policy built around human ownership, human dignity, and human leverage. Otherwise “dominance” will mean what it usually means: a small number of institutions gaining power by treating the rest of the country as expendable. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit jamesfeltonkeith.substack.com

    5 мин.
  2. 8 мар.

    The $1,000,000 DEI Insurance Problem

    For International Women's Day I sat down with Kit Chaskin and Lauren Golanty from Third to First for a conversation on the Inclusionism podcast about a novel insurance solution to an work place conflict problem that we have quietly lived with for decades: learn how we disincentivize workplace conflict. At first the discussion sounded like a conversation about HR systems. But the more we talked, the clearer it became that the real issue is much deeper than HR. It is about how the insurance system itself is structured. And once you see the structure, it becomes obvious why so many workplace problems end up costing companies hundreds of thousands—or even millions—of dollars. Because the system is built to respond after the damage is done. Most workplace lawsuits do not begin dramatically. They begin with moments that seem small at the time. An employee raises a concern. A manager brushes it aside. HR logs the complaint but decides the issue does not require escalation. Weeks pass. Sometimes months. Then a lawyer’s letter arrives. By the time that happens, the problem has already become expensive. To understand why, it helps to understand one of the most basic concepts in insurance: the difference between first parties, second parties, and third parties. In any insurance policy, the first party is the one who buys the coverage. In Employment Practices Liability Insurance—often called EPLI—that first party is the employer. The second party is the insurance company providing the coverage. The third party is someone who claims they were harmed. In workplace liability cases, that third party is typically an employee alleging discrimination, harassment, retaliation, or wrongful termination. EPLI is designed as third-party liability insurance. That means the policy activates only when the third party—the employee—files a legal claim against the first party—the employer. In other words, the insurance system only turns on once the workplace conflict becomes a legal dispute. Which is precisely the moment when the problem has become the most expensive. Most executives think employment lawsuits are rare events. But insurers know something different. Across large workforces, employment complaints are statistically predictable. Insurance underwriters often estimate that roughly half a percent to one percent of employees will file some form of workplace complaint each year. It sounds like a small number until you apply it to a real company. A firm with 1,000 employees can expect somewhere around five to ten complaints annually. A company with 5,000 employees might see twenty-five to fifty complaints each year. For organizations employing 10,000 people or more, the number can exceed fifty or even a hundred workplace complaints in a single year. Most of those issues never become lawsuits. Many are resolved internally, and some move through regulatory channels like the EEOC, which receives roughly 70,000 employment discrimination charges every year in the United States. That was before Trump destroyed the agency. But a small percentage escalate. And when they do, the cost curve changes dramatically. Legal defense costs alone often exceed $200,000, even when the company ultimately wins the case. Settlements frequently average between $75,000 and $125,000, while jury verdicts can climb into the $500,000 to $1 million range. Large verdicts—sometimes called “nuclear verdicts”—can exceed $5 million, and occasionally much more. Once you factor in turnover, leadership distraction, lost productivity, and reputational damage, the real cost of a serious workplace dispute often lands somewhere between one and two million dollars. The strange part is that many of those cases begin with problems that could have been addressed much earlier—and much more cheaply. This is where the idea behind Third to First becomes interesting. When Kit and Lauren described their approach, they framed it as a simple shift in perspective. Instead of waiting for employees to become third-party legal claimants, organizations should treat workplace conflict as a first-party operational risk. That may sound like subtle insurance language, but the financial difference is enormous. If a workplace issue is addressed early—through mediation, investigation, or corrective action—the cost might fall somewhere between $10,000 and $50,000, depending on the size of the organization and the complexity of the situation. If that same issue escalates into litigation, the cost can quickly reach $300,000 to $1 million or more. Preventing even one lawsuit can save hundreds of thousands of dollars. Preventing several per year can save millions. The deeper reason this matters today is that companies themselves have changed. Fifty years ago, most corporate value came from physical assets: factories, equipment, inventory. Today, more than ninety percent of the value of companies in the S&P 500 comes from intangible assets—human capital, intellectual property, brand trust, and organizational culture. See my previous article on the 90% valuation of people. When workplace systems fail, companies do not just face legal exposure. They damage the very assets that create long-term value. For decades the insurance industry has quietly paid for the consequences of workplace failures. Employment Practices Liability Insurance exists because organizations sometimes fail to manage people risk effectively. But what the conversation with Kit and Lauren made clear is that the real opportunity may not lie in paying those claims. It may lie in preventing them. If workplace conflicts can be addressed before employees become legal adversaries, insurance stops functioning purely as a financial backstop. It becomes part of a system that encourages organizations to fix problems before they turn into million-dollar disputes. And once you see the math behind workplace risk, the logic becomes hard to ignore. Most million-dollar lawsuits begin as ten-thousand-dollar problems. The real question is whether companies choose to solve them early—or wait until the insurance policy activates. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit jamesfeltonkeith.substack.com

    48 мин.
  3. 27 февр.

    Brittany Kaiser & JFK on Data as Property

    Own Your Data: From Scandal to Structure A fireside conversation between James Felton Keith and Brittany Kaiser When Joel Telpner opened the evening by jokingly asking the audience to hand over their driver’s licenses and Social Security numbers, the room laughed. It was a sharp reminder of how casually we’ve surrendered personal information for years. But beneath the humor was a serious premise: data has moved from the margins of business and politics to the center of both. And the question is no longer whether it is valuable. The question is who owns it—and who gets paid. This conversation was not about personalities or partisan politics. It was about structure. The goal was to go deeper than headlines and public relations and push into the mechanics of ownership itself. From Ad Tech to Political Shock When Brittany and I first met in November 2016 at the Cambridge Analytica office, the conversation was about business models and the ad-tech ecosystem. At the time, data targeting was standard practice. Advertising had long operated on behavioral insights. Politics was simply catching up. Outrage only surfaced when data use became visibly political—even though corporations had been exploiting personal data for years. That asymmetry revealed something deeper: the public didn’t object to data extraction itself. They objected when they saw its power. Her departure from Cambridge Analytica wasn’t rooted in naïveté about data. She had worked in political campaigns since 2007–2008, including the Obama campaign, where social media data first became a serious political tool. Data wasn’t new. What was new was the scale, opacity, and consolidation of control. The “walled garden” model—where platforms closed APIs not to protect users but to centralize monetization—made that consolidation explicit. The model is simple:Users generate the value.Platforms capture the revenue.Institutions sell access to audiences. And individuals receive “free” services in exchange. Data as Property, Not Just a Byproduct The response was the hashtag #OwnYourData. But slogans aren’t solutions. The real shift comes when the idea is grounded in property rights. The argument is pragmatic: in American law, nothing is more protected than property. If personal data is treated as property, then established legal frameworks—transparency, consent, licensing, compensation—can apply. A simple analogy makes it clear: Airbnb. Before someone uses your house, they disclose who they are, what they’ll do, how long they’ll stay, and how much they’ll pay. You consent, and you’re compensated. Why shouldn’t data operate the same way? Who wants it? What will they do with it? For how long? What is the compensation? This reframes data from a passive byproduct into an asset. Data as Labor and Economic Participation My position intersects with—but also extends beyond—property rights. I treat data not only as property, but as productive input. I approach problems as an engineer. Most problems are distribution problems. And the current economy has a distribution issue: productivity has increased dramatically, but wages have not scaled with it. If individuals are generating data constantly—through work, consumption, communication, and engagement—then they are participating in production. Yet they are not compensated for that participation. Universal Basic Income attempts redistribution from the top down. I have argued instead for what I call Universal Basic Ownership. Ownership scales. Welfare programs do not. If individuals own their data—both data they generate and data derived about them—they gain an economic stake in the systems they fuel. That changes participation from passive extraction to active transaction. The Literacy Gap A recurring theme in the conversation was disclosure versus comprehension. Terms and conditions technically represent “informed consent.” But no one reads them. And even fewer understand them. Even legislators questioning tech executives often lack fundamental data literacy. If lawmakers don’t grasp the mechanics, how can consumers? The problem is not merely transparency. It is intelligibility. Without data literacy, consent becomes procedural rather than meaningful. Licensing in Reverse One of the more practical ideas raised was reversing the licensing model. Instead of users accepting platform terms, platforms would license data from users under clearly defined conditions. That would mean: Limited purpose Time-bound use Explicit compensation Enforceable constraints Blockchain and smart contracts were discussed as technical mechanisms to encode such agreements. Whether blockchain is the ultimate solution is secondary. The core idea is contractual symmetry. Right now, contracts are one-directional. That imbalance is structural. Inequality and Valuation A fair concern was raised: if data is monetized, will some people’s data be worth more than others’? The answer is uncomfortable: yes, in some markets it already is. But transparency cuts both ways. If value disparities exist, surfacing them allows for legal and policy remedies. Hidden extraction prevents accountability. Ownership does not automatically produce equality. But opacity guarantees inequity. From Awareness to Market Shift Five years ago, cybersecurity products were difficult to sell because people didn’t perceive risk. Today, breaches are common knowledge. Language shapes markets. Once “data ownership” enters the public lexicon, business models adapt. The question for executives is not whether change is coming. It is whether they will build systems that: Share value, or Continue extracting until regulation forces recalibration. The Core Shift Data has been described as: The new oil The new currency The new asset class All are partially correct. But the more precise framing may be this: Data is productive human output in a digital economy. If it is productive, it generates value.If it generates value, ownership matters. This was not a call for privacy panic. It was a call for structural clarity. We are no longer debating whether data is powerful. We are debating whether individuals will remain raw material—or become economic participants. That is not a technical question. It is a political and legal one. And it is now part of the lexicon. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit jamesfeltonkeith.substack.com

    39 мин.
  4. 21 февр.

    Data Theft: Diversity Photos vs Adobe Inc.

    Gerald Carter states on the podcast that his company, Diversity Photos, filed claims against Adobe Inc. after alleging that Adobe used their licensed images to train its generative AI model, Firefly, beyond the scope of their agreement. The dispute centers on three core allegations described in the episode: * Breach of contractCarter argues the license granted to Adobe was for promoting and distributing their work — not for AI model training. Adobe reportedly relied on language allowing “new features and services” to justify use in Firefly. * Copyright-related claimsIncluding direct and contributory copyright infringement, particularly around AI training use. * Negligence / metadata strippingCarter claims images were made available online without watermark or copyright metadata, enabling downstream AI systems (e.g., third-party models) to ingest them. The case did not proceed in court in the traditional sense. The contract required arbitration, and according to Carter: * Most claims were dismissed in arbitration. * One negligence claim survived initially. * The process became cost-prohibitive (projected ~$100K). * The matter concluded without prejudice due to inability to continue funding arbitration. * Adobe allegedly attempted a discretionary “bonus” payment (~$1,100) for dataset usage, which Carter rejected. A Forbes article by Jasmine Browley. That piece frames the broader tension between creators who built digital businesses under one set of platform norms and the shift toward AI systems trained on that content. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit jamesfeltonkeith.substack.com

    1 ч. 24 мин.
  5. DEI = AI

    18 февр.

    DEI = AI

    We are proud to announce IncluionScore.AI which is the result of about 40,000 and 45 years of DEI&B know-how. Inclusion Score is the tech arm of the ISO-30415 Certifying Body and our Chair, James Felton Keith is expanding on the original certification course from University of Georgia to train everyone on how to implement DEI with rigor while filling in all of their knowledge gaps. --- The ISO-30415 AI Has Arrived For years, I’ve traveled the world — from Denmark to Durban to Detroit — speaking on panels about Diversity, Equity, Inclusion, and Belonging. And everywhere I go, I hear the same thing: But that’s not true. The problem isn’t that DEI hasn’t been defined.The problem is that the people who helped define it can’t be everywhere at once. So we built something that can. For years, when organizations wanted to align with ISO-30415 — the International Standard for Diversity & Inclusion — we would run diagnostics. We would audit. We would walk companies step-by-step through governance, HR, product delivery, and supplier diversity. It worked. But it didn’t scale. So we took our ISO-30415 manual.We took the standard itself.And we built an AI trained on more than 44,000 incidents, methodologies, workbooks, court dockets, depositions, and case studies spanning over 45 years — from late 1980 through today. From Title VII challenges…To supplier diversity frameworks…To global governance models…To implementation playbooks. We trained the system on it all. And now it lives at: InclusionScore.ai The biggest gap in the DEI profession isn’t passion. It isn’t intent. It isn’t even resistance. It’s communication. In 2021, global consensus was reached around ISO-30415. That consensus clarified what DEI is and what it is not. DEI is not: Just HR Just supplier diversity Just governance Just product strategy It is the integration of all four: Governance Human Resources Product & Service Delivery Supplier Diversity When those four systems operate together under a structured maturity model — that’s Diversity & Inclusion Service Management. That’s ISO-30415. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit jamesfeltonkeith.substack.com

    3 мин.

Об этом подкасте

James Felton Keith explores the tenets of inclusionism . As always, JFK is joined by Peter Willumsen and friends. His ethic of inclusionism is at the core of his work on global diversity equity inclusion and belonging. JFK is the Founder of Techstars Venture Backed, InclusionScore Companies and a lecturer at the University of Georgia's Terry College of Business. In 2017 he became the first Black LGBT person to run for federal office in the United States during his bid for Congress to represent Harlem. jamesfeltonkeith.substack.com