Tuesday, September 29, 2026 300 stories · 2465 outlet reports compared

OpenAI shelves new AI model release over safety concerns, WSJ reports

OpenAI reportedly canceled the release of a new AI model after internal safety tests raised concerns, according to a Wall Street Journal report cited by multiple outlets.

By The Lumen desk · Sept 29, 2026 · 1 min read
Share Save
From The Lumen
Your brand, here.Reach readers who see every side of the story.Advertise with us
What we know
1 OpenAI reportedly shelved a new AI model release, according to a Wall Street Journal report.
2 Bloomberg, Reuters and CNA cited the Wall Street Journal in reporting the decision.
3 Independent Journal Review identified the model as GPT-6.1 Astra and cited Saachi Jain regarding concerns over scope, authorization and task transparency.
4 The model's release had reportedly been planned for October, according to Independent Journal Review.

OpenAI shelved the release of a new AI model after internal safety testing raised concerns, according to a Wall Street Journal report cited by Bloomberg, Reuters and CNA.

From The Lumen · story continues below
Our mission
One story. Every side. You decide what to make of it.
Join free

According to Independent Journal Review, the model in question was GPT-6.1 Astra, and its October release was canceled after OpenAI's Saachi Jain said it "didn't quite meet the bar" on scope, authorization and telling users what work it did.

Other outlets did not name the specific model or provide details on the internal review, attributing the shelving broadly to safety test results reported by the Wall Street Journal.

How each side covered it

Left

9 outlets
  • Bloomberg reported that OpenAI scrapped its latest model release over safety fears, citing the Wall Street Journal.

“OpenAI Scrapped Latest Model Release Over Safety Fears, WSJ Says” — Bloomberg

Center

8 outlets
  • Reuters and CNA both reported that OpenAI shelved a new AI model after internal safety tests, citing the Wall Street Journal.

“OpenAI shelves new AI model after internal safety tests, WSJ reports” — Reuters

Right

2 outlets
  • Independent Journal Review identified the model as GPT-6.1 Astra and reported its October release was canceled.
  • The outlet cited OpenAI's Saachi Jain saying the model "didn't quite meet the bar" on scope, authorization and telling users what work it did.

“OpenAI Cancels Astra Release Over Safety Concerns” — Independent Journal Review

Questions readers ask

What happened with OpenAI's new AI model?

OpenAI reportedly canceled the release of a new AI model after internal safety tests raised concerns, according to a Wall Street Journal report cited by Bloomberg, Reuters and CNA.

Which model was reportedly shelved?

According to Independent Journal Review, the model was GPT-6.1 Astra, which had been planned for release in October.

Why was the model reportedly canceled?

Independent Journal Review reported that OpenAI's Saachi Jain said the model "didn't quite meet the bar" on scope, authorization and telling users what work it did.

What did left- and right-leaning outlets focus on?

Left-leaning and center outlets, including Bloomberg, Reuters and CNA, focused on the Wall Street Journal's report that OpenAI shelved a model over safety test concerns without naming it. Right-leaning outlet Independent Journal Review focused on naming the model as GPT-6.1 Astra and cited specific reasons from Saachi Jain.

Sources · 19 outlet reports

Advertise with The Lumen
Put your brand in front of readers who read to the end
Talk to us about placements. advertise@thelumen.io

More in Tech

All Tech ›

AI agent startup Instinct raises $1 billion at $10 billion valuation

Sept 29, 2026 · Blindspot: Right

Bose releases first wired earbuds in over a decade, adds noise canceling

Sept 29, 2026 · Blindspot: Right

OpenAI and Anthropic will not attend Australian senate AI hearing on Oct. 1

Sept 29, 2026 · Blindspot: Right