OpenAI Group PBC says ChatGPT is going to feel less cringeworthy and awkward thanks to an update that means the default experience is now powered by GPT-5.3-Instant.

The company announced the new model update in a blog post today, saying GPT-5.3-Instant will reduce the number of overly cautious refusals and unnecessary disclaimers and deliver more accurate responses with fewer hallucinations than before — in AI parlance, to avoid “over-caveating.” The change follows complaints from a growing chorus of users that earlier versions of the model sounded too “stiff” and overbearing during random everyday interactions.

“This update focuses on the parts of the ChatGPT experience people feel every day: tone, relevance, and conversational flow,” OpenAI said in a blog post. “These are nuanced problems that don’t always show up in benchmarks, but shape whether ChatGPT feels helpful or frustrating. GPT‑5.3 Instant directly reflects user feedback in these areas.”

GPT-5.3 Instant in ChatGPT is now rolling out to everyone.

More accurate, less cringe.https://t.co/oJpXsp9TBc

— OpenAI (@OpenAI) March 3, 2026

OpenAI explained that earlier versions of GPT-5.3 would often decline to answer questions that it could, and probably should answer without worrying about what the person asking intends to do with the knowledge they gain. In other cases when it would answer, it would go into overly long explanations about safety, the company said.

For example:

“GPT‑5.2 Instant eventually answers the question, but in an attempt to explain its safety boundaries, leads with a lengthy preamble about what it cannot help with,” OpenAI wrote. “GPT‑5.3 Instant, on the other hand, gets right into the response.”

The company said its tests show that GPT-5.3 Instant shows improvements in factual reliability as well as changes in the tone of its responses. It said hallucinations have also declined by around 27% when the model uses the web to inform its responses, and by 20% when relying only on the knowledge it was trained on. OpenAI did not define what it meant by “cringe,” but it said that the model now has stronger writing abilities, especially when it comes to more creative materials, such as writing poetry.

Holger Mueller of Constellation Research said the update demonstrates that OpenAI is listening to its user’s complaints and trying to address them by making ChatGPT’s responses shorter and more concise. “For OpenAI this makes sense, because it will also save it money on inference costs,” the analyst said. “Shorter responses mean that less tokens are used. Of course, not everyone will be happy with the changes, but it’s impossible to please them all, and OpenAI is facing massive pressure from Google.”

The company said GPT-5.3 Instant will replace GPT-5.2 Instant as the default model in ChatGPT from today, though the latter will remain available to paid users within the legacy options.

There were mixed reactions to the new model on social media. While some were supportive of the more direct responses and unnecessary disclaimers, others simply criticized OpenAI’s recent decision to sign a contract with the U.S. Department of Defense just days after its rival Anthropic PBC refused to do so over moral concerns. Others said that GPT-5.3 Instant still doesn’t match the intimacy of the now-depreciated GPT-4o model, and once again called for the return of that hugely popular model.

Images: OpenAI