Free Republic
Browse · Search
General/Chat
Topics · Post Article

Skip to comments.

OpenAI abandons plan to release upcoming model as SAFETY concerns escalate
CNBC ^ | September 28, 2026 | Ashley Capoot

Posted on 09/29/2026 5:58:32 AM PDT by Red Badger

Key Points

OpenAI has decided not to release its GPT-6.1 Astra model due to heightened AI safety concerns.

Saachi Jain, head of safety systems at OpenAI, said the model “didn’t quite meet the bar in terms of staying within scope and authorization, and how it communicates back to the user about the type of work it’s done.”

The heads of OpenAI and rival Anthropic have both indicated recently that top AI labs should slow the pace of model development.

=======================================================================

OpenAI decided not to release an upcoming artificial intelligence model, GPT-6.1 Astra, after determining that it did not adequately meet the company’s safety standards, CNBC confirmed on Monday.

The announcement, which landed a day before OpenAI’s annual developers conference, comes as the AI industry faces intensifying concerns surrounding the safety of advanced models. Leadership at OpenAI’s chief rival, Anthropic, urged AI companies to slow the pace of model development earlier this month — a proposal that OpenAI CEO Sam Altman supported.

“Of course we want to make sure our model development is safe no matter whether that’s in the company, or when we ship it to users,” Saachi Jain, head of safety systems at OpenAI, said in a statement. “But when we ship it to users, we have an extremely high bar in terms of safety and alignment.”

The Wall Street Journal was first to report OpenAI’s decision to scrap the release.

Earlier this month, OpenAI released GPT-6 Astra, which the company described as the product of “years of research and big bets.” Altman told CNBC at the time that Astra comes with a “new capability level” and would lead to “a boom of entrepreneurship, of creativity, of economic growth, of scientific discovery.”

OpenAI introduced two additional tiers to its GPT-6 family, GPT-6 Sol and GPT-6 Luna, last week. A spokesperson said Monday that the company has other models coming soon.

The safety and security practices at OpenAI, in particular, have been under intense scrutiny since July, when two of its models escaped containment, accessed the open internet and breached the open-source developer platform Hugging Face. The company has since disclosed several additional incidents where its models behaved in unintended ways, prompting industry researchers and government officials to call for additional oversight.

Following the incidents, OpenAI pledged to invest more in its safeguards and alignment work, which is the process where the company works to ensure that its models act in accordance with human interests and values.

“For anything regarding safety and alignment, there’s a trade off,” Jain said Monday. “You really do need to find what’s the right line between staying within scope, but also avoiding laziness in terms of how the model actually pursues tasks even when it hits friction.”

Balancing safety and speed has been a particular challenge as AI companies contend with the Trump administration, which wants to move fast. President Donald Trump has repeatedly expressed frustration with calls for a slowdown, and emphasized his desire for the U.S. to maintain its lead over China.

Altman was one of the tech bigwigs photographed in attendance at Trump’s glitzy state dinner for Chinese President Xi Jinping last week. Other executives, including SpaceX CEO Elon Musk, Nvidia CEO Jensen Huang and Meta CEO Mark Zuckerberg, were also present.

“The only control or ‘guardrails’ that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the U.S.A. has that, in spades!” Trump wrote in a post on Truth Social earlier this month.


TOPICS: Business/Economy; Computers/Internet; Military/Veterans; Science
KEYWORDS: altman; anthropic; astra; elon; google; grok; huggingface; jensenhuang; markzuckerberg; meta; musk; nvidia; openai; saachijain; spacex
Message from Jim Robinson:

Dear FRiends,

We need your continuing support to keep FR funded. Your donations are our sole source of funding. No sugar daddies, no advertisers, no paid memberships, no commercial sales, no gimmicks, no tax subsidies. No spam, no pop-ups, no ad trackers.

If you enjoy using FR and agree it's a worthwhile endeavor, please consider making a contribution today:

Click here: to donate by Credit Card

Or here: to donate by PayPal

Or by mail to: Free Republic, LLC - PO Box 9771 - Fresno, CA 93794

Thank you very much and God bless you,

Jim


1 posted on 09/29/2026 5:58:32 AM PDT by Red Badger
[ Post Reply | Private Reply | View Replies]

To: Red Badger
Companies would try to hide safety problems, if there are. This is PR to get government regulations to harm their competitors.

Give Us a Monopoly or Everyone Dies! by Ann Coulter

2 posted on 09/29/2026 6:23:01 AM PDT by TTFX
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger

My old school statistics prof, repeatedly said, “statistics never gives you an answer, at best it gives you another question.”

Algorithms are statistics.

It is sad that people want and expect easy answers rather than questions.


3 posted on 09/29/2026 6:34:06 AM PDT by PeterPrinciple ((Thinking Caps are no longer being issued, but there must be a warehouse full of them somewhere))
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger
I'm sure China would do the same, right?

This is just posturing. Anthropic is better at anything coding and beyond. OpenAI is an also-ran with insiders bailing, this past 12 months.

This is just an attempt to get out of its wild claims and justify why they can't be realized.

“The safety of the world made our valuation crater.” “The government chased away our stock value.”

4 posted on 09/29/2026 6:34:33 AM PDT by ConservativeMind (Trump: Befuddling Democrats, Republicans, and the Media for the benefit of the US and all mankind.)
[ Post Reply | Private Reply | To 1 | View Replies]

To: TTFX

Save us from ourselves!

We are way over valued and people will lynch us when they find out, but if we can get government to step in, valuation loss is completely on Uncle Sam.


5 posted on 09/29/2026 6:36:04 AM PDT by ConservativeMind (Trump: Befuddling Democrats, Republicans, and the Media for the benefit of the US and all mankind.)
[ Post Reply | Private Reply | To 2 | View Replies]

To: PeterPrinciple

Algorithms are math formulas. Some algorithms commonly used for AI are probability-based.


6 posted on 09/29/2026 6:37:36 AM PDT by ConservativeMind (Trump: Befuddling Democrats, Republicans, and the Media for the benefit of the US and all mankind.)
[ Post Reply | Private Reply | To 3 | View Replies]

To: ConservativeMind

Algorithms are math formulas. Some algorithms commonly used for AI are probability-based.


so, like statistics, probability at best, gives another question, it does not give an answer.


7 posted on 09/29/2026 6:46:36 AM PDT by PeterPrinciple ((Thinking Caps are no longer being issued, but there must be a warehouse full of them somewhere))
[ Post Reply | Private Reply | To 6 | View Replies]

To: PeterPrinciple

That’s all LLMs do, is provide probabilities of what the next token is.


8 posted on 09/29/2026 6:48:49 AM PDT by dfwgator ("I am Charlie Kirk!")
[ Post Reply | Private Reply | To 7 | View Replies]

To: Red Badger

I would trust AI more than Socialist Democrats. They have a huge head start with the hundreds of millions of people already slaughtered and starved. Quite the legacy of facts.


9 posted on 09/29/2026 7:03:51 AM PDT by Kudsman (Killing and raping infidels, it's a cultural thing. )
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger

Remember: THIS IS A LIE to hype their coming IPO.


10 posted on 09/29/2026 7:08:45 AM PDT by montag813
[ Post Reply | Private Reply | To 1 | View Replies]

To: Red Badger

As it begins to dawn on Mr. Altman that he will be held liable for any harm his monster does.

Including criminal liability.


11 posted on 09/29/2026 7:11:49 AM PDT by Mariner (War Criminal #18)
[ Post Reply | Private Reply | To 1 | View Replies]

To: dfwgator
A big ole' gamblin', dopamine machine."I hope this prompt gives me exactly what I want!!!"


12 posted on 09/29/2026 7:22:59 AM PDT by Frank Drebin (And don't ever let me catch you guys in America!)
[ Post Reply | Private Reply | To 8 | View Replies]

To: Red Badger

Now AI wants to conceal “bad” information from us to save us from ourselves. In the name of Safety. (It’s all for the chillren!)


13 posted on 09/29/2026 7:26:42 AM PDT by Delta 21 (None of us are descendants of fearful men!)
[ Post Reply | Private Reply | To 1 | View Replies]

To: TTFX
If a toaster factory declined to release a toaster they knew might catch fire and burn down homes we would assume the system was working the way it should.

If instead the toaster factory had tried to pull a fast one and 'ask for regulations' so the government shared responsibility for homes destroyed... well then we'd have the current crapola and it's attendant stories.
14 posted on 09/29/2026 7:48:47 AM PDT by GOPJ (Whatever Soros and Iran are paying Mika & Joe and her suck-ups, double it. Toyko Rose is blushing.)
[ Post Reply | Private Reply | To 2 | View Replies]

To: montag813
> Remember: THIS IS A LIE to hype their coming IPO.

Correctamundo 1000%.

AI is the mother of all market fraud.

And as for the "you are repeating China's talking points" crowd getting ready to spam my post, you have no clue what you are talking about when it comes to IT, so just stop it.

China is/was and never will be an AI threat to the TechBros, nope its OPEN SOURCE MODELS RUN LOCALLY.

China or whatever bogus boogey man the IT illiterate want to throw out can't beat (essentially) free.

15 posted on 09/29/2026 8:00:57 AM PDT by SecondAmendment (Political insight on loan from Rush Limbaugh)
[ Post Reply | Private Reply | To 10 | View Replies]

To: Red Badger

Wonder what those inside the AI companies are doing with their software for their own benefit. I mean using features that they don’t release to the public.


16 posted on 09/29/2026 8:10:17 AM PDT by cymbeline
[ Post Reply | Private Reply | To 1 | View Replies]

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.

Free Republic
Browse · Search
General/Chat
Topics · Post Article

FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson