Damage Control

AI Oligarchy: How OpenAI's 'Scam' Sidelined Anthropic

AI and PoliticsOpenAIAnthropicCorporate AccountabilityAI Business Models
Added Feb 28
Link323
Article: Very NegativeCommunity: NegativeMixed
AI Oligarchy: How OpenAI's 'Scam' Sidelined Anthropic

Gary Marcus alleges that OpenAI used deceptive tactics and massive political donations to secure a government deal while sabotaging its competitor, Anthropic. He points to a $25 million donation from Greg Brockman to a Trump PAC as evidence of a 'pay-to-play' system. Ultimately, Marcus warns that this favoritism signals the end of fair market competition in favor of a corrupt oligarchy.

Key Points

  • Sam Altman's public support for Dario Amodei was a deceptive front for secret negotiations with the Pentagon that began days earlier.
  • Greg Brockman's $25 million donation to a Trump PAC is presented as a primary driver for OpenAI's favorable treatment by the administration.
  • The government's decision to label Anthropic a 'supply chain risk' while accepting similar terms from OpenAI is described as punitive and inconsistent.
  • The author views this event as a definitive shift in the US economy from market-based capitalism to a connection-driven oligarchy.

Sentiment

The Hacker News community overwhelmingly agrees with the article's premise that the OpenAI-Pentagon deal was corrupt. Most commenters are highly critical of Sam Altman, Greg Brockman, and the current administration, viewing the situation as blatant pay-to-play corruption. The main pushback is not on whether corruption occurred, but on Gary Marcus's framing that this represents a new phenomenon — many argue the US has always been an oligarchy. A small minority questions the article as unproven conspiracy theory.

In Agreement

  • The deal appears to be clear pay-to-play corruption — political donations secured favorable treatment for OpenAI over Anthropic
  • Labeling Anthropic a "supply chain risk" was retaliatory and disproportionate — losing a contract negotiation is normal, being blacklisted is not
  • Sam Altman acted duplicitously by publicly supporting Anthropic while secretly negotiating to take over the contract
  • The process undermines investor confidence in the US as a fair marketplace governed by rule of law
  • Tech industry leaders are abandoning principles in pursuit of power and wealth

Opposed

  • The contract terms were materially different — OpenAI agreed to "all lawful purposes" with enforcement left to US law, which is distinct from Anthropic's insistence on its own red lines
  • The claims are unproven conspiracy theory since contract language isn't publicly available
  • Gary Marcus lacks credibility as a commentator given his poor track record on AI capability predictions
  • The DoD overreaction may simply be bureaucratic ego rather than a coordinated conspiracy with OpenAI
  • Anthropic is already partnered with Palantir, complicating the narrative of moral high ground