Price:

AI & TECH

Enterprises reject frontier AI labs over data leakage fears

Sunday, July 5, 2026 · from 3 podcasts
Stacker News LiveAll-InThis Week in AI
  • Enterprises are building their own AI to stop labs from stealing proprietary data and competing in their verticals.
  • AI job loss data is flawed; companies investing heavily in AI are actually hiring more people.
  • Frontier labs lobby for safety regulations that kill open-source competition and create a duopoly.

The trust between enterprises and frontier AI labs has evaporated. Companies no longer believe they can safely rent intelligence from OpenAI or Anthropic. On All-In, David Sacks pointed to Anthropic’s launch of Claude Design as a direct assault on its partner Figma - a clear signal that labs will use customer data to identify and dominate high-value markets.

The fear is data leakage and intellectual property theft. Alex Karp argues that outsourcing core operations to general AI consensus is dangerous. David Friedberg notes that life sciences companies are rejecting Anthropic’s proposals to share proprietary data, fearing the commoditization of their core assets. The model is broken.

"If you build on their platform, you are essentially providing the R&D for your future competitor."

- David Sacks, All-In with Chamath, Jason, Sacks & Friedberg

The response is sovereign AI. Palantir and Nvidia formed a partnership where government agencies will own the hardware, data, and model weights. Chamath Palihapitiya argues that using a proprietary control plane with open-source models is already 16 times cheaper for enterprise tasks like code migration. The value is shifting from raw tokens to data retention and on-premise execution.

This sovereign shift coincides with a regulatory clampdown that further centralizes power. On Stacker News Live, Carl stated that Anthropic will introduce KYC for Claude starting July 8th, flagging accounts and restricting access. Frontier capabilities are being reserved for the state while the public receives nerfed versions.

"If the government hoards elite models like 'Mythos' - which can allegedly penetrate classified U.S. systems in hours - the public remains vulnerable to zero-day exploits that only the state can identify."

- Host Keon, Stacker News Live

The narrative of mass AI-driven unemployment isn’t showing up in the numbers. Friedberg highlighted a study of 21,000 U.S. firms: companies spending the most on AI grew their headcount by 10% over two years. Entry-level hiring rose by 12%. AI spend signals business expansion, not staff culling.

Displacement is coming, but slowly. Jason Calacanis predicts significant job loss in customer support, data entry, and driving within 10 years. For now, the market is creating more jobs around AI integration. Friedberg argues that as automation scales, ‘human in the loop’ will become a premium service.

Regulation is the final battleground. David Sacks argues that Anthropic and OpenAI are attempting to create a regulatory duopoly by framing frontier models as potential ‘cyber weapons.’ This invites government intervention that restricts open-source alternatives. The recent flip-flop on Anthropic’s export restrictions - where the Commerce Department lifted them after the lab replaced its lead negotiator - shows that ‘safety’ is often a proxy for political alignment.

The sovereign path is clear: enterprises will own their models, governments will hoard the elite tools, and the public will get a restricted, verified version of yesterday’s intelligence.

#Big Tech#Regulation#Agents
AnthropicNvidiaOpenAIPalantirClaudeFigma

Source Intelligence

- Deep dive into what was said in the episodes

Stacker News Live
Stacker News Live

SNL #230: Broke into almost all US classified systems, not in weeks, in hours • Jul 3

  • Carl describes Peter Todd delivering a Hilux truck and Starlink internet to Ukrainian troops, documenting drone attacks, glide bombs, and soldiers using basement restaurants.
  • Carl says Anthropic will introduce KYC for Claude starting July 8th, flagging accounts and restricting access after a 30-day window.
  • Carl argues KYC for AI models creates friction but most users will comply, pushing only a tiny fraction towards alternative, less advanced open-weight models.
  • Carl states Mythos AI model broke into classified U.S. systems within hours, finding critical vulnerabilities and showing a leap in cyber exploit capabilities.
  • Carl claims frontier AI labs will stop releasing advanced models publicly if governments mandate special access, stalling public innovation and creating a security gap.
Also from this episode: (10)

Protocol (9)

  • Carl says working in Bitcoin during a bear market is hard and financially volatile, often tied to Bitcoin price, contrasting with soul-crushing but lucrative fiat jobs.
  • Carl argues PubLab is a crucial community accelerator for entering the Bitcoin ecosystem, enabling people to meet others, start careers, and companies.
  • Carl speculates Bitcoin's feisty community and lack of women might be due to fewer resources and comforts for women in the ecosystem.
  • Carl believes Bitcoin's bear markets wash out fake participants, forcing them to lose interest until prices rise again.
  • Carl says pacing yourself and avoiding high-engagement media channels like X, which amplify conflict and AI hype, preserves energy and passion for Bitcoin.
  • Carl states raw avocado analyzed Hal Finney’s logs to conclude Finney was the first Bitcoin user after Satoshi, using node discovery via IRC to rule out Dustin Trammell.
  • Carl says Greg Maxwell submitted a 17-page FCC comment opposing KYC mandates for phones, arguing it harms security, privacy, and free speech.
  • Carl states Justin Moon's C-Link project offers a modular Lightning wallet and node solution for seamless Bitcoin payments, with a new website and designer.
  • Carl describes LDK's async payments feature, allowing Lightning payments when a node is offline, requiring both sender and receiver to use an LSP.

Culture (1)

  • Carl analyzes a resurgence in movie theater attendance, attributing it to post-COVID social reconnection, a recession-driven entertainment surge, and simply better films.
#Models#Regulation#Macro#Safety
All-In with Chamath, Jason, Sacks & Friedberg
All-In with Chamath, Jason, Sacks & Friedberg

AI Sovereignty Wars, Palantir-Nvidia Deal, SCOTUS Birthright Ruling, Newsom's CA Budget Lie • Jul 3

  • Palantir and Nvidia formed a sovereign AI partnership, with Palantir using Nvidia's Nemotron open models to build a custom frontier model for the U.S. government. Government agencies will own the hardware, data, and model weights.
  • Alex Karp argues that enterprises are uncomfortable with frontier AI labs due to distrust and the risk of losing proprietary knowledge. He believes outsourcing core operations to general AI consensus is dangerous.
  • Chamath's company, 8090, found that wrapping an open-source model with their software factory was 16.4x cheaper than using Anthropic Opus 48 alone, despite being three times slower. Using their harness with Claude was 1.4x cheaper and 1.5x faster.
  • Friedberg notes that life sciences companies are rejecting Anthropic's proposals to share proprietary data, fearing commoditization of their core assets. He predicts a shift towards enterprises training and running their own models on local hardware.
  • Jason warns founders against partnering with major AI labs like OpenAI, asserting that such platforms historically acquire or compete directly with their partners, risking the partner's business.
  • Sacks explains that the temporary export control on Anthropic's Fable 5 was due to Dario Amodei's claims of Mythos being a 'cyber weapon,' Amazon's report of guardrail failures, and Dario's initial refusal to roll back the model.
  • Sacks argues that once a model is open source, it ceases to be tied to its country of origin, as it can be forked and run on local, secure hardware. Banning open source models in the U.S. would isolate the country and increase costs.
  • Sacks cites a study by Ramp and Ravello Labs of over 21,000 U.S. firms showing that high AI adopters grew faster and increased overall headcount by 10% and entry-level headcount by 12% over two years.
  • Jason predicts significant job displacement in customer support, data entry, and driving within 10 years, citing advances in AI and robotics like Figure and Optimus, especially for package sorting and last-mile delivery.
  • California's state budget increased 65% from $215 billion in 2019 to $355 billion today. The state's revenue heavily depends on personal income tax, with the top 1% (150,000 people) paying half ($70 billion of $142 billion).
  • Since 2019, at least 15 Fortune 500 companies and 2,100 mid-to-large companies have exited California due to high corporate tax rates. An average of 1-1.5% of the state's adjusted gross income leaves annually, predicting a 15% loss over a decade.
  • California is implementing an 8% sales tax on software, projected to generate $1 billion annually, and a new $2 billion annual tax on healthcare insurance. The state also made the 13.3% top income tax bracket, enacted in 2012, permanent and increased it to 14.4%.
  • California faces projected $40 billion annual budget deficits by 2028-29. The state has $1.4 trillion in public debt, $664 billion in unfunded pension liabilities (potentially $1.5 trillion), and $175 billion in retiree healthcare obligations.
  • Friedberg predicts that California's financial crisis will lead to calls for a federal bailout, potentially triggering red states to question their participation in the union if asked to cover blue states' liabilities. Chamath foresees a default on pensions, a new state constitution, and a political 'red wave.'
Also from this episode: (6)

Enterprise (1)

  • Sacks explains that Anthropic vertically integrates by launching its own apps like Claude Design and Claude Code, after observing the success of companies building on its models. This strategy is compared to Microsoft's or Google's historical market dominance.

Markets (1)

  • Sacks highlights Anthropic's launch of Claude Design blindsiding its partner Figma, leading to a 50% drop in Figma's stock while Anthropic's valuation surged.

Immigration (3)

  • SCOTUS struck down Trump's executive order to end automatic birthright citizenship for children of illegal immigrants and temporary visa holders, with a 6-3 or 5-4 ruling.
  • Friedberg suggests birthright citizenship should apply to children of legal U.S. residents, but not temporary visitors. Sacks emphasizes that the 14th Amendment's original intent was to grant citizenship to freed slaves.
  • Chamath argues that Western countries risk losing cultural identity due to immigration and emphasizes the importance of assimilation. Friedberg advocates for an immigration policy that prioritizes 'makers' who contribute economically over 'takers' who rely on social benefits.

Macro (1)

  • California Governor Newsom announced a $351 billion 'balanced' budget through 2028, but Friedberg critiques it as relying on accounting tricks, debt, and new taxes, rather than genuine revenue-expense alignment.
#Enterprise#Models#Regulation#Labor#AI Infrastructure
This Week in AI
This Week in AI

Mythos, Sonnet 5, GLM-5.2 Dominate the News Cycle | Episode 20 • Jul 2

  • Andrew Burman argues US AI policy, including blocking models like GPT-5.6 and restricting access, is harming US technological supremacy by enabling Chinese competition. He believes open-source models are being distilled from US models and the US should push ahead faster.
  • Victor Perez states that creative AI models like Korea's face lower regulatory risk than cybersecurity-focused models, citing NSFW images and impersonation as primary concerns rather than national security threats.
  • Dave Gar explains action models differ from language models by being designed for task execution, such as operating computers, automating work, or acting as personal assistants.
  • Andrew Burman notes enterprises strongly desire open-source models like GLM 5.2 for cost reduction, despite wanting access to powerful models like Fable, creating a cost-performance trade-off dilemma.
  • Andrew Burman describes trusted access programs from Anthropic (Project Glasswig) and OpenAI, which allow large companies and cybersecurity firms to evaluate vulnerabilities in models like Fable or GPT-5.6 before public release.
  • Victor Perez explains Korea AI released raw model variants like K2 RAW to improve tunability and research, arguing overly post-trained models narrow output possibilities and hinder innovation.
  • Dave Gar says AGI Inc.'s on-device action models initially used open-source LLMs but now train custom models, collecting millions of data points from paid Android screen recordings.
  • Victor Perez observes a shift in AI focus from pure capability to performance (speed, cost), citing examples in creative AI where models like Nano Banana reached a 'minimum viable intelligence' threshold.
  • Andrew Burman states enterprises face a dramatic increase in AI inference costs, citing a customer whose agent harness consumed 75% of its 2026 budget in a weekend.
  • Dave Gar predicts a bifurcation in AI: super-intelligence for hard sciences like research, and smaller specialized 'good enough' models for everyday tasks like ordering coffee.
  • Victor Perez claims Higsfeld AI employs aggressive, sometimes deceptive marketing tactics, using AI-generated videos to exaggerate product capabilities, and questions the accuracy of reported revenue figures.
  • Andrew Burman says Run Layer's revenue grew 8x since signing its Series A term sheet, serving customers in tech, financial services, and healthcare.
  • Victor Perez acknowledges distillation of model IP, particularly from Chinese labs, is a real threat to companies like Anthropic, but Korea encourages it as distilled models remain inferior to original ones.
  • Andrew Burman reports Run Layer's average monthly AI spend per employee is between $5,000 and $7,000, with top spenders reaching up to 50% of their salary in token costs.
  • Dave Gar defines AGI as nearly achieved for most tasks, with ASI arriving around 2030, and says his p(doom) is decreasing due to responsible development and AI's role in eliminating tedious work.
  • Victor Perez struggles to define AGI but suggests it could arrive via improved model context and tool access rather than raw intelligence. His p(doom) decreases as open-source closes the capability gap.
Also from this episode: (1)

Agents (1)

  • Andrew Burman describes agent swarms as multiple parallel agents enabling superhuman productivity, currently prevalent in coding but expanding to sales, marketing, and design functions.
#Models#Regulation#Startups#Agents#Chips

Related Stories