AI Business Models
How AI companies make money, including subscriptions, usage pricing, marketplaces, services, margins, distribution, and defensibility.
Cost Per Token Is Replacing FLOPS as the AI Infrastructure Metric
Shruti Koparkar of NVIDIA’s Accelerated Computing team argues that AI infrastructure should be evaluated by token economics rather than by GPU-hour pricing or FLOPS per dollar. On NVIDIA’s AI Podcast, she lays out a four-part framework — token utility, supply, demand and monetization — in which cost per token becomes the central measure of business value. Koparkar says NVIDIA Blackwell’s system-level design delivers 50 times more tokens per watt than Hopper and 35 times lower token cost, while lower token costs will expand GPU demand by making more AI workloads economically viable.
Ivan Zhao Says AI Makes Companies Flatter, Not Hierarchy-Free
Notion founder and CEO Ivan Zhao argues that AI will not make companies hierarchy-free, but can reduce the amount of human routing that makes hierarchy slow. In a conversation with Brian Halligan, Zhao describes Notion’s answer as “jazz mode”: a deliberately decentralized company that still has structure, but relies on high-agency people, ex-founders and model-enabled teams to improvise as product and market conditions change. His broader case is that AI-era leaders have to refound around the technology itself, not just bolt it onto the old SaaS operating model.
Kled Founder Alleges Luel Copied Its Human Data Marketplace
This Week in Startups put two founder arguments side by side: Mercury chief executive Immad Akhund said the fintech’s new $200mn round is meant to create strategic flexibility for a profitable company seeking a bank charter, while Kled founder Avi Patel argued that an alleged copycat in the human-data marketplace category threatens trust in a business built on consent and compliance. Jason Calacanis treated Patel’s dispute with Luel, Y Combinator and General Catalyst less as an intellectual-property case than as an ethics and diligence signal for investors.
Agent-Native Clouds Need Faster Primitives, Not New Ones
Railway founder Jake Cooper argues that software infrastructure does not need to abandon its old primitives for agents, but must make them much faster, cheaper, safer and more observable. In a wide-ranging interview with swyx and Alessio, Cooper lays out Railway’s attempt to build an agent-native cloud through own-metal data centers, production forks, progressive rollouts and deployment loops that assume thousands of concurrent software-producing actors rather than one human pushing a pull request.
Generative AI’s Revenue Stack Is Still Inverted Toward Chips
Stanford adjunct lecturer and Altimeter partner Apoorv Agrawal argues in MS&E435 that generative AI’s economics still look unlike the software and cloud cycles investors often use to value it. In his estimates, AI revenue has grown sharply, but gross profit remains concentrated in semiconductors, while applications face inference costs, thin monetization and uncertain paths to mass-market utility. The question he puts to students is not whether AI demand exists, but how long the stack’s inverted shape can persist before applications and infrastructure capture more of the value.
TSMC’s Wafer Scarcity May Be Preventing an AI Overbuild
Investor Gavin Baker argues on Invest Like The Best that the AI boom is being organized less by software adoption than by scarcity: compute demand is outrunning power, wafers, and frontier-model access. In his account, Anthropic’s growth, Nvidia’s position, TSMC’s capacity discipline, and even SpaceX’s possible orbital compute are all expressions of the same constraint. Baker’s central claim is that the AI cycle may avoid a classic infrastructure bubble only if physical bottlenecks, especially leading-edge wafer supply, keep capital from building far ahead of demand.
Modern AI Needs Inference and Incentives, Not AGI Framing
Michael I. Jordan argues that modern AI is being framed around the wrong object: an isolated intelligent machine rather than the collective economic systems in which machine-learning components actually operate. In this conversation, the Berkeley statistician and computer scientist says AGI is mostly a PR term, and that the field’s harder problems lie in inference, uncertainty, incentives, markets, and mechanism design. His case is not that recent models are unimpressive, but that prediction and fluent language are only pieces of systems that must be engineered around human institutions.
Google Turns TPU Capacity Into a Blackstone-Backed Neocloud
Bloomberg Technology’s Caroline Hyde and Ed Ludlow frame Google’s new venture with Blackstone as an attempt to turn Google’s TPU capacity into an AI cloud business outside Google Cloud. Bloomberg Intelligence’s Mandeep Singh argues the structure could help Google meet external demand for its chips by shifting more of the data-center burden to Blackstone, creating a TPU-based rival to Nvidia-centered neocloud providers.
Parallel Launches Marketplace to Pay Publishers for AI Agent Work
Parallel founder and former Twitter CEO Parag Agrawal argues that AI agents are breaking the web’s existing content economics by using publisher and creator material to perform valuable work without tying compensation to that value. In a Bloomberg Technology interview, Agrawal said Parallel’s new Index marketplace is meant to pay publishers, data providers, and independent creators according to their content’s measured contribution to an agent’s completed task, rather than through ads, subscriptions, citations, or flat licensing deals.
ElevenLabs Adds Albert Einstein’s Voice to Its Licensed AI Marketplace
ElevenLabs is offering a licensed AI version of Albert Einstein’s voice through its Iconic Marketplace, positioning it for narration, education, documentaries, and immersive storytelling. The company argues that Einstein’s voice can be used as both a cultural artifact and a creative tool, while saying the marketplace is curated and that each voice is approved and managed with the relevant rights holder.
AI Backlash Reaches Commencement as Graduates Face a Reshaped Job Market
Jason Calacanis and Alex Wilhelm argue that the boos greeting pro-AI commencement speeches are a visible sign of AI’s legitimacy problem with new graduates entering the workforce. On This Week in Startups, they frame the reaction less as technophobia than as distrust: students have already seen AI weaken academic norms, threaten entry-level work, concentrate wealth around frontier labs, and expand systems of surveillance and data capture. Their discussion returns to a central question: whether workers, founders, consumers, and citizens have any meaningful control over the AI systems now reshaping their choices.
AI’s Demo Phase Is Giving Way to Infrastructure and Compliance Fights
On Diet TBPN, John Coogan and Jordi Hays framed the day’s AI news around the point where software claims meet physical, financial and political constraints. Coogan argued that the Sanders-AOC data center proposal is less a simple moratorium fight than a question of definitions, grid costs and who pays for externalities, while Hays said local objections cannot simply be dismissed. Across segments on ChatGPT personal finance, circular revenue, office prompting, Tesla’s lead and a possible SpaceX IPO, the show treated AI’s next phase as an institutional test rather than a demo problem.
Economic Entanglement, Not Decoupling, Defines the New China Bargain
Salesforce CEO Marc Benioff joined the All-In hosts for a discussion that framed U.S.-China relations, enterprise AI, and the software selloff around the same question: when dependence is a stabilizer and when it becomes leverage. Benioff argued that more trade with China can lower conflict risk and that large software platforms remain valuable because AI still needs trusted customer data, cash-flowing distribution, and enterprise deployment. David Friedberg, Chamath Palihapitiya, and Jason Calacanis extended the argument across Taiwan, chips, AI assistants, El Niño-driven food risk, and private-market SPVs, where interconnection can either absorb shocks or transmit them.
AI Software Winners Will Own Context, APIs, or Outcomes
Tasklet chief executive Andrew Lee argues that AI software is consolidating toward a few horizontal agent platforms that hold context, connect tools, generate interfaces, and choose among models. In a discussion with Nathan Labenz, Lee says Tasklet has rewritten its agent stack around file-system memory, agentic search, and provider-specific context management because the chat transcript is no longer enough. He also frames Anthropic as both Tasklet’s critical supplier and a major competitor, making model neutrality central to Tasklet’s bid to survive the AI transition.
Figma Says AI Makes Design More Valuable as Code Gets Easier
Figma CEO Dylan Field told Bloomberg that the company’s stronger-than-expected quarter shows AI is expanding rather than undermining its market. He argued that as large language models make code easier to generate, design becomes the more valuable layer above it — while acknowledging that AI features carry real inference costs that Figma is now trying to monetize through usage credits.
Self-Driving Startups Shift From Science Risk to OEM Deployment
Wayve chief executive Alex Kendall and Waabi chief executive Raquel Urtasun argue that self-driving has moved from a basic research problem to an execution problem built around end-to-end AI, world models, OEM partnerships and deployment economics. In this This Week in Startups discussion, Kendall makes the case for licensing Wayve’s “intelligence layer” across consumer vehicles and robotaxis, while Urtasun says Waabi’s L4-native Driver-as-a-Service model can scale first through trucking and then robotaxis. Both reject the idea that autonomy is simply solved, but they present the remaining challenge as integration, validation, regulation and commercialization rather than a missing scientific breakthrough.
AI’s Value Is Moving From SaaS Margins to Hardware Capacity
PwC technology, media and telecommunications leader Dallas Dolen argues that the AI boom is a real infrastructure and business-model shift, but one constrained by chips, construction labor, telecom capacity, copper, power and enterprise economics. In a PwC-sponsored interview, he says value is moving from SaaS toward hardware, software margins are compressing, and most companies are less limited by compute access than by token costs, security rules and measurable return on investment. Dolen’s view of enterprise AI is practical and bounded: agents are working in defined back-office, sales and legal tasks, while broader automation will depend on cost, governance and human oversight.
OpenAI Prepares Legal Action as Apple Partnership Falls Short
Bloomberg’s Mark Gurman reports that Apple’s partnership with OpenAI has deteriorated because OpenAI expected deep ChatGPT integration across Apple software and a multibillion-dollar annual opportunity, but received a narrower set of features. Gurman says OpenAI has tried to renegotiate, believes talks have stalled, and is preparing possible legal action while still seeking an out-of-court resolution. Apple has not commented, but Gurman says it has its own concerns about OpenAI’s privacy practices, durability, leadership, and recruitment from Apple hardware teams.
Oura Seeks Clinical Validation for Longer-Term AI Health Prediction
Oura chief executive Tom Hale told Bloomberg Technology that the company’s AI work is not a new response to the current market cycle but an extension of years of prediction work in wearables. His argument is that Oura can move from near-term wellness signals, such as illness or menstrual-cycle alerts, toward longer-range health guidance, provided the science and regulatory validation support it. Hale said the company is still stopping short of diagnosis while it works with the FDA, including on blood-pressure submissions, and framed Oura’s hardware as an advantage in an AI market where software is easier to copy or generate.
Cerebras Raises $5.55 Billion in Year’s Biggest IPO
Cerebras chief executive Andrew Feldman used the AI chipmaker’s $5.55 billion IPO to argue that public investors are valuing the company as a fast-inference infrastructure supplier, not merely another semiconductor listing. In a Bloomberg Technology interview before trading began, Feldman said demand is concentrated around speed, claimed Cerebras is about 15 times faster than its nearest competitor, and pointed to large relationships with OpenAI and AWS as evidence of commercial traction, while acknowledging that the AWS agreement is still being finalized.
Pax Silica Aims to Secure the Full AI Supply Chain
U.S. Under Secretary of State for Economic Affairs Jacob Helberg argues that AI dominance depends on securing the full industrial supply chain behind compute, not just advanced semiconductors. In an interview with Sarah Guo and Elad Gil, Helberg presents Pax Silica as a 14-country economic-security coalition meant to build commercially viable allied supply-chain platforms, starting with a 4,000-acre industrial zone in the Philippines. He frames the strategy as a private-sector-led alternative to China’s Belt and Road model, combining domestic reindustrialization with partner-country specialization in critical inputs such as minerals, robotics components, and processing capacity.
AI Is Forcing Startups to Return Capital or Rebuild Around Agents
AI is forcing founders and investors to make decisions faster than venture’s last cycle assumed they would have to, Jason Calacanis, Alex Wilhelm, Jenny Fielding, Dave McClure and Sam Lessin argue on This Week in Startups. Fielding’s example is a legal-tech founder who raised a $15mn Series A and, six months later, planned to return the money because he believed Claude and other models could erode the company’s long-term value. The same pressure is showing up in private markets, where demand for exposure to OpenAI and Anthropic is straining company controls over secondary sales, SPVs and liquidity.
OpenAI and Anthropic Are Compressing the Market for Thin AI Wrappers
Sam Lessin of Slow Ventures argues that OpenAI and Anthropic are moving into the application layer fast enough to threaten many AI startups built as thin wrappers on foundation models, while Jenny Fielding and Dave McClure contend that workflow depth, distribution and niche focus may still protect some companies. The broader debate links that pressure to a weak secondary market, a doubtful 2026 IPO rescue and a venture model Lessin says must shift away from multi-stage capital deployment toward early, priced exposure to scarce founder talent.
Compute Allocation Is Anthropic’s Core Constraint as Claude Revenue Surges
Anthropic CFO Krishna Rao argues that the company’s rise is best understood through compute: a scarce capital asset that must be bought years ahead and constantly reallocated across model training, customer demand, internal automation and future products. In an interview with Patrick O’Shaughnessy, Rao says ordinary forecasting and software-margin frameworks break down when model capability, adoption and revenue compound together, leaving Anthropic to manage growth through scenarios rather than point estimates.