OpenAI launches GPT‑5.3 Instant amidst backlash over Pentagon agreement

OpenAI launched its new GPT-5.3 Instant model, promising more accuracy and “less cringe,” even as the company faces backlash over its agreement with the U.S. Pentagon.

OpenAI launched its new GPT-5.3 Instant model, promising more accuracy and “less cringe,” even as the company faces backlash over its agreement with the U.S. Pentagon.

The ChatGPT-maker said that the GPT-5.3 Instant model would have fewer unnecessary refusals and “preachy” disclaimers, while also delivering sharper contextualisation, better understanding of question subtext, and a more consistent response tone while chatting.

“GPT‑5.3 Instant significantly reduces unnecessary refusals, while toning down overly defensive or moralizing preambles before answering the question,” said OpenAI in a blog post about the release, adding, “GPT‑5.3 Instant also improves the quality of answers when information comes from the web. It more effectively balances what it finds online with its own knowledge and reasoning—for example, using its existing understanding to contextualize recent news rather than simply summarizing search results.”

The company further promised a reduction in hallucination with the user-feedback evaluation showing that hallucinations decreased by 22.5% with web use and by 9.6% without web access.

OpenAI’s GPT‑5.3 Instant is available to all users in ChatGPT.

However, the release comes during a challenging time for OpenAI’s reputation.