GPT-5 is getting panned on Reddit as Sam Altman preps for AMA

GPT-5 is out, the early opinions are in, they usually’re not nice.

Many ChatGPT followers have taken to Reddit and different social media platforms to precise their frustration and disappointment with OpenAI’s latest basis mannequin, launched on Thursday.

A fast glimpse of the ChatGPT subreddit (which isn’t affiliated with OpenAI) exhibits scathing opinions of GPT-5. For the reason that mannequin started rolling out, the subreddit has stuffed with posts calling GPT-5 a “catastrophe,” “horrible,” and the “greatest piece of rubbish whilst a paid consumer.”

Awkwardly, Altman and different members of the OpenAI crew are anticipated to do a Reddit AMA at 2:00 p.m. ET at the moment, the place they’re going to reply questions on GPT-5. Already, the questions have piled up in anticipation of the Q&A. Many customers are demanding that OpenAI convey again GPT-4o as an alternative choice to GPT-5.

What Redditors are saying about GPT-5

Most of the unfavourable first impressions say GPT-5 lacks the “persona” of GPT-4o, citing colder, shorter replies. “GPT-4o had this… heat. IT was witty, inventive, and surprisingly private, like speaking to somebody who bought you. IT didn’t simply spit out solutions; IT felt like IT listened,” mentioned one redditor. “Now? Every thing’s so… sterile.”

One other said, “GPT-5 lacks the essence and soul that separated Chatgpt (sic) from different AI bots. I sincerely want they carry again 4o as a legacy mannequin or one thing like that.”

A number of redditors additionally criticized the truth that OpenAI did away with the choice to decide on completely different fashions, prompting some customers to say they’re canceling their subscriptions. “I awoke this morning to seek out that OpenAI deleted 8 fashions in a single day. No warning. No alternative. No “legacy possibility,” posted one redditor who mentioned they deleted their ChatGPT Plus account. One other consumer posted that they canceled their account for a similar purpose.

As Mashable reported yesterday, GPT-5 integrates numerous OpenAI fashions into one platform, and ChatGPT will now select the suitable mannequin based mostly on the consumer’s immediate. Clearly, some customers miss the outdated system and fashions.

Sarcastically, OpenAI has additionally drawn criticism for having too many mannequin choices; GPT-5 was speculated to resolve this confusion by streamlining the earlier fashions beneath GPT-5.

Expectations for GPT-5 couldn’t have been greater — and which may be the actual downside with GPT-5.

Gary Marcus, a cognitive scientist and creator identified for his analysis on neuroscience and synthetic intelligence — and a widely known skeptic of the AI hype machine — wrote on his Substack that GPT-5 makes “Good progress on many fronts” however disappoints in others. Marcus famous that even after multi-billion-dollar investments, “GPT-5 shouldn’t be the large leap ahead folks lengthy anticipated.”

The final time OpenAI launched a frontier mannequin was over two years in the past with GPT-4. Since then, a number of opponents like Google Gemini, Anthropic’s Claude, xAI’s Grok, Meta’s Llama, and DeepSeek R1 have caught as much as OpenAI on benchmarks, comparable agentic options, and consumer loyalty. For a lot of, GPT-5 had the ability to bolster or topple OpenAI’s reign because the AI chief.

With this in thoughts, IT‘s inevitable that some customers could be disillusioned, and lots of ChatGPT customers have shared constructive opinions of GPT-5 as nicely. Time might blunt these criticisms as OpenAI makes enhancements and tweaks to GPT-5. The corporate has additionally traditionally been aware of consumer suggestions, with Altman being very energetic on X.

“We at present imagine the easiest way to efficiently navigate AI deployment challenges is with a good suggestions loop of speedy studying and cautious iteration,” the corporate’s mission statement avows.

Disclosure: Ziff Davis, Mashable’s father or mother firm, in April filed a lawsuit towards OpenAI, alleging IT infringed Ziff Davis copyrights in coaching and working its AI programs.