Tag: chatgpt

The much touted feature last showed up as «naughty chats» in the Android app just two weeks ago, but is now postponed indefinitely.

It was supposed to debut in the first quarter of 2026, after first being conceived as «letting adults be adults» in October 2025. It was supposed to run in tandem with age verification, rolled out this January.

OpenAI says that «we still believe in the principle of treating adults like adults, but getting the experience right will take more time,» writes Axios.

The postponement is due to the company focusing on other high priority issues, such as «intelligence, personality improvements, personalization,» and «making the experience more proactive.»

Read more: Axios and TechCrunch.

The new Thinking and Pro models are more «capable and efficient» and is the first OpenAI model with native computer use skills. It also improves on hallucinations and Office files creation — areas where Anthropic has been thriving.

— Together with advances in general reasoning, coding, and professional knowledge work, GPT‑5.4 enables more reliable agents, faster developer workflows, and higher-quality outputs across ChatGPT, the API, and Codex, OpenAI writes.

On hallucinations, it is 33% less likely to be wrong in its responses, and 18% less apt to have mistakes in replies compared to GPT‑5.2.

Continue reading “OpenAI’s new ChatGPT-5.4 has native computer use and less hallucinations”

Hallucinations are down 27% when it uses the web and ~20% in reasoning, and OpenAI did actually notice that the 5.2 model often cames across like a nanny — berating users with obvious, condescending lectures for harmless questions.

This has been solved by reducing unnecessary «dead ends,» caveats and «declarative formulations» in order to let the conversation flow more freely — solving one of many users’ major gripes with the current OpenAI models.

The model is also better at search, combining search responses with its own knowledge and reasoning. This should put results into context instead of just summarizing results.

Bu the real update with this model is bringing ChatGPT to a more natural style and tone with freely flowing conversation — which should be far less «cringe.»

OpenAI is promising less «Stop. Take a breath»-responses, or «I’m going to calm this down» kinds of outputs. It should also be better at creative writing.

GPT-5.3-instant is available today in the app (choose Instant in the model selector), API and web, and updates to Thinking and Pro are «coming soon.»

Read more: OpenAI’s announcement, writeups on 9to5Google, TechCrunch and The Register.

The new ChatGPT Android app (v1.2026.055) mentions a new "Naughty chats" setting

"This setting lets ChatGPT use spicier, adult-themed language, when you ask. Available to users 18+ only." pic.twitter.com/7oZiAGbWJ4

— Tibor Blaho (@btibor91) February 26, 2026

According to ChatGPT watcher Tibor Blaho, adult mode is getting closer to going live, with «spicier, adult-themed language.»

The idea was to get age verification on ChatGPT sorted first, so teens could be protected from «sensitive content,» while the app would let «adults be adults.»

Verification was rolled out in January, and people have been asking when ChatGPT for adults would finally ship.

ChatGPT can be useful, but also carries some risk — especially when it’s browsing the web or using apps.

That’s why OpenAI is now launching «an advanced, optional setting for higher risk users.»

The idea is to disable tools an adversary might use to «to exploit to exfiltrate sensitive data,» such as prompt injection attacks in files or on the web.

When using Lockdown Mode, web use is limited to cached content only, and no live network access is available outside of OpenAI’s «controlled network»

Most people will never need this feature, OpenAI says, but it might be useful for corporate execs, security teams, journalists and officials.

OpenAI is also launching an «Elevated Risk»-setting, that lets users fine-tune network access for Atlas and Codex.

Both features are coming to users «in the coming months.»

Read more: OpenAI’s announcement.

While touting a return to more than 10% monthly growth for ChatGPT and «insane» Codex growth, Sam Altman also said they are preparing to launch «an updated Chat model» this week, writes CNBC.

That would likely be GPT-5.3, debuted last week for Codex. It’s a blazingly fast and more capable successor to 5.2, and is said to be one of the first models used to make itself.

The new Codex model has been a great success, funneling a 60% growth in overall use just last week.

That model is also available to Free and Go users for a limited time, and is now being extended with perhaps reduced limits, Altman says:

— We want everyone to be able to try Codex and start building.

Read the full scoop on CNBC

Brought back by popular demand after the turbulent release of GPT-5, OpenAI is now sending off the GPT‑4o, GPT‑4.1, GPT‑4.1 mini, OpenAI o4-mini and GPT-5 models on February 13th.

The reason for bringing them back was how clinical and cerebral, and less friendly GPT-5 had gotten, leading to a very public backlash.

After this, OpenAI spent a lot of time thinking about personality and customization — leading to GPT-5.2, which now has the overwhelming majority of users.

The previously very popular 4o model is now only used by 0.1% of users.

In their announcement today, OpenAI also says they are making progress towards a more creative version of GPT for adults, having rolled out age prediction earlier this month.

Read more: OpenAI’s announcement, discussion on r/ChatGPT.

Scientists are often plagued by having to use different platforms and apps for seemingly mundane things.

That’s where OpenAI’s new service comes in — as a LaTeX-native interface, it combines the paper writing in the same space as it does equations, references and «surrounding context.»

Of course there is a GPT-5.2 engine right there, so it’s easy to cross-reference and check for originality, find citations and proofread — and it’s a fairly nice research assistant, to boot.

Prism is available as of now for anyone who has a personal account with OpenAI.

Read more: OpenAI’s launch page, launch tweet, The tool itself, writeup on Gizmodo.

Last month offered a sneak peek at OpenAI’s support page for «age prediction,» and now they are ready to go live.

Age prediction works by taking in several factors such as usage time, account age and the time the user is active.

It is developed «in dialogue» with experts from the American Psychological Association, ConnectSafely, and Global Physicians Network⁠.

After several wrongful death lawsuits involving teens, disturbing mental health conversations and stating that principles are in conflict, OpenAI was more or less forced into doing something on both teen use and parental controls.

Those deemed under 18 by the AI will be served a vanilla version of ChatGPT without beauty standards, violence or romantic role-play, to name a few examples.

If someone gets wrongfully lobbed into the teen experience, and it does make mistakes, they can upload a selfie for proper verification through Persona.

Read more: OpenAI’s announcement, writeups on Reuters, CNBC, Engadget.

ChatGPT Go was first launched as a cheap way to access more GPT queries in India in August, 2025, for around $5. Now, the subscription has become popular enough to expand globally to every market where OpenAI is active.

The idea is that for a just little more money, $8 in the USA, you can get ten times the queries, ten times the file use, and ten times the image generations compared to the free plan.

OpenAI says that since the introduction of the Go subscription tier, they have seen extensive everyday use in tasks such as writing, learning, image generations and problem-solving.

Along with the Free tier, ChatGPT Go will be showing ads once they are ready.

Read more: OpenAI’s announcement, OpenAI’s sub overview. Writeups on The Verge, MacRumors.