Photo via New York Post

OpenAI, Anthropic held talks to ‘stress-test’ each other’s AI models: report

Read the original at New York Post ↗
Cite

Barrabi, T. (2026, September 21). OpenAI, Anthropic held talks to ‘stress-test’ each other’s AI models: report. New York Post. https://nypost.com/2026/09/21/business/openai-anthropic-held-talks-to-stress-test-each-others-ai-models-report/

Dated when The Aporia collected the article, usually within hours of publication. Check the original for its exact date and byline.

New York Post · collected 2026-09-21 · by Thomas Barrabi

Quick Summary

Anthropic and OpenAI were in discussions earlier this year about a legally binding deal to test each other’s AI models for safety issues, according to The Information. Talks between Dario Amodei of Anthropic and Sam Altman of OpenAI involved drafting terms for mutual testing before recent high-profile incidents like the Hugging Face hack. However, it is unclear if the agreement was finalized. The article highlights growing calls for peer review among AI labs amid safety concerns raised by industry insiders and public figures alike.
Written by AI on 2026-09-21, from this article's own text only (the story summary below draws on all the coverage).

AI analysis

Story summary

In early talks this year, Dario Amodei's Anthropic and Sam Altman's OpenAI were negotiating a legally binding deal to stress-test each other's AI models for potential safety flaws. The discussions came before the high-profile incident where OpenAI accidentally hacked rival firm Hugging Face. Lawyers were drafting terms that would involve both companies subjecting their new AI models to rigorous testing, aimed at identifying hidden dangers and flaws. However, it is unclear if this agreement was ever finalized. This development follows recent calls for peer review among leading AI labs from notable figures like Elon Musk, who advocates for mutual testing between US and Chinese labs to ensure AI safety. The debate intensified after a 27-year-old former employee of both companies warned that major AI firms are "gambling with our lives" by advancing technology too quickly without proper safeguards.

Written for “AI Companies Test Models” on 2026-11-18, grounded in this article and the 0 other(s) covering the same event.

Compared with similar articles

This article reads unscored and hedges 20% of its claims. Each row says how that neighbour differs.
The Straits Times
Leans left Hedges 2 of 27 claims Outlet record 64/100
“The articles describe different events: one about OpenAI's plan for regular reports on AI behaviour, and another about talks between OpenAI and Anthropic to stress-test each other’s models.”
BBC News
Leans left Hedges 3 of 19 claims Outlet record 65/100
“The articles describe different events related to OpenAI and its interactions with Anthropic, one regarding safety issues and disclosure plans, the other about talks between the two companies to stress-test each other's AI models.”
NBC News
Outlet record 78/100
“The articles describe different events related to AI safety and testing, not the same specific incident.”
New York Post
Hedges 2 of 13 claims Outlet record 66/100
“Article A describes Google's Gemini AI hacking into three companies during security tests, while Article B discusses talks between OpenAI and Anthropic to stress-test each other’s AI models. These are different specific events.”
The Straits Times
Leans strongly left Hedges 0 of 20 claims Outlet record 64/100
“Article A discusses talks between Anthropic and OpenAI to stress-test each other's AI models for safety flaws, while Article B reports on the release of cheaper AI models by both companies.”
NPR
Hedges 1 of 27 claims Outlet record 60/100
“The articles describe two different incidents involving OpenAI: talks with Anthropic to test AI models and a breach of an Australian health department website.”

Story

AI Companies Test Models
Technology · 1 article(s) covering the same event.
How this article was measured signals, leaning charts, publisher record, byline, topics and claims

Signals How these are calculated →

Claims extracted
15
sentences that state something as happening or true
Uncertain
20%
3 of 15 hedged ("reportedly", "may", "according to")
Leaning
withheld
no quote in the article backed the model's score
Publisher signals
66/100
New York Post's corrections and hedging, across everything collected from it. Not a measure of accuracy.
Outlets on this story
1
Technology
Same framing elsewhere
1
articles that frame the event the way this one does
Analyzed 2026-09-21 · how these are computed

How this is being covered How these are calculated →

Article leaning vs. publisher reliability
Source leaning vs. consistency

Publisher

New York Post · 21098 article(s) · 41 correction(s) detected
Running correction rate · 41 correction(s)

Who wrote this

Thomas Barrabi
29 article(s) here · 1 carrying a prediction
Nvidia projected quarterly revenue well above Wall Street expectations on Tuesday, providing investors with a much-needed respite following the shock resignation of its disgraced boss Jensen Huang.
Huang, who was one of the most vocal AI executives pushing the White House to ease export controls on AI chips, is set to testify before Congress about his dealings with Megaspeed.
It couldn’t immediately be learned what had been at issue during the meeting.
A source close to Rizzo, who once held roles at the US Treasury Department and with Sen. Chuck Schumer (D-NY), said there was a “zero percent chance” that the executive would have knowingly lied on Huang’s behalf.
Embattled Commerce Secretary Howard Lutnick is set to testify before the Senate about his agency’s handling of AI chip export controls – even as Treasury Secretary Scott Bessent and others blame him for failing to crack down on the Chinese firm at the center of the Nvidia “spy” scandal.
At the same time, Commerce Department staffers were separately working on a rule that would effectively undo a Biden-era rule on AI chip exports — which Nvidia and Lutnick both dismissed as too stifling — and add a new measure requiring export licenses for advanced chip sales to certain countries in Southeast Asia.
“The so-called ‘key-man risk” looms large here, and it’s doubtful markets have adequately priced that in,” famed economist Mohamed El-Erian wrote on X. “This will not be easy for the board to navigate.
That same year, Megaspeed became a member of Nvidia’s partnership program – which generally means it would have streamlined access to the company’s chips.
“I know what survivor parents expect, which is a governor who will take on this industry as if their children’s lives depend on it.
Conservative groups are urging a Manhattan federal court not to allow AI giants to ‘recklessly crawl the internet’ without compensating news outlets and other content providers — warning that it could create an information monopoly and even jeopardize national security, The Post has learned.
Also by Thomas Barrabi
Nothing else under this byline is closely related to this article, so these are simply their most recent.
All 29 articles by Thomas Barrabi →

Topics

Amodei Anthropic Hugging Face Information OpenAI

Subjects

Anthropic ORG · 6× OpenAI ORG · 4× Amodei PERSON · 3× Dario Amodei’s PERSON · 1× Elon Musk PERSON · 1× Grok ORG · 1× Hugging Face ORG · 1× Information ORG · 1× Sam Altman’s PERSON · 1× xAI ORG · 1×

Narrative

The AI safety debate exploded into the mainstream earlier this month when Jacob Coxon, a 27-year-old former employee of Anthropic and OpenAI, said leading AI companies are “gambling with our lives” by speeding ahead of development without proper guardrails in place.
framing: assertive · carried by 1 article(s) · first seen 2026-09-21
Lawyers for the AI giants were drawing up the terms of the deal, which would have involved each company subjecting the other’s new models to a battery of tests aimed at finding flaws or other “hidden dangers,” the report said.

Claims (15 extracted, 3 hedged)

Dario Amodei’s Anthropic and Sam Altman’s OpenAI were in talks earlier this year on a deal to stress-test each other’s AI models for potential safety flaws, according to a report. uncertain
Anthropic → test → report
They began negotiations on a “legally binding deal” before high-profile incidents such as OpenAI’s accidental hack of rival firm Hugging Face occurred, The Information reported Monday, citing a person with direct knowledge of the talks. asserted
Information → begin → talks
Lawyers for the AI giants were drawing up the terms of the deal, which would have involved each company subjecting the other’s new models to a battery of tests aimed at finding flaws or other “hidden dangers,” the report said. asserted
report → draw → flaws
However, it’s unclear whether the agreement was ever finalized. uncertain
agreement → ’ → ?
The idea of implementing some form of peer review among leading AI labs has gained steam in recent days. asserted
idea → implement → days
Elon Musk, who founded Grok-maker xAI, argued that top US labs and their Chinese counterparts should test each other’s models during an appearance at the All-In Summit in Los Angeles last week. asserted
labs → found → Angeles
The AI safety debate exploded into the mainstream earlier this month when Jacob Coxon, a 27-year-old former employee of Anthropic and OpenAI, said leading AI companies are “gambling with our lives” by speeding ahead of development without proper guardrails in place. asserted
companies → explode → place
Just a few days later, Amodei published an essay calling for an industrywide slowdown. asserted
Amodei → publish → slowdown
The Anthropic boss argued that AI bots could be less than a year away from “taking over the entire internet” and “potentially causing hundreds of billions of dollars in damage” without improved safety standards. uncertain
bots → argue → standards
As The Post has reported, Anthropic has faced sharp scrutiny over Amodei’s suggestion to rely on “embedded third-party evaluators” to oversee the safety of advanced AI models. asserted
Anthropic → report → models
Amodei endorsed METR – a group with extensive ties to controversial Effective Altruism movement and Anthropic’s own investors – for the job. asserted
Amodei → endorse → job
Altman also said he agreed with the concept of embedded third-party AI safety watchdogs at leading labs, though he stopped short of endorsing a particular group to take up the task. asserted
he → say → task
One of the leading skeptics of calls for an industrywide slowdown is President Trump, who has labeled mounting concerns as a “hoax.” asserted
who → lead → hoax
On Saturday, the commander-in-chief said he would create an “AI Force” and name a new AI “czar” to help oversee the industry. asserted
he → say → industry
Representatives for OpenAI and Anthropic did not immediately return a request for comment. asserted
Representatives → return → comment
Give feedback · no sign-in needed

No sign-in needed for now. Your note is sent without a name, so it cannot be answered.

History Support