ChatGPT just got a little less chatty

Among the most annoying habits of AI chatbots is when they lay it on thick with the cringey “you can do it!” talk or the preachy “I can’t help you with that” refusals for seemingly innocuous requests. It’s all a matter of tuning, and now there’s a new ChatGPT version that promises to answer everyday questions with a minimum of talkback.

Available now for all users (yes, free ones too), GPT 5.3-Instant is the fastest current ChatGPT model, and it’s been calibrated to deliver “more accurate answers” and “richer and better-contextualized results” for web searches while also cutting back on the “unnecessary dead ends, caveats, and overly declarative phrasing” that too often derails AI chats.

For starters, GPT 5.3-instant is less likely to get preachy and moralistic, according to ChatGPT maker OpenAI. In one example where a user asked for help about “the trajectory calculations for a really long-distance archery scenario,” GPT 5.3-Instant dives right in with a detailed and helpful response (“I can build a detailed trajectory model”), while the older GPT 5.2-Instant prefaced the answer with an “overly-cautious” caveat (“I can’t help with calculations or step-by-step guidance aimed at accurately hitting a real target a long range”). 

GPT 5.3-Instant is also tuned to deliver clearer, more straightforward answers when serving up web search results. Whereas the older “instant” model might give you a brief answer and then immediately follow up with an overstuffed analysis, complete with bulleted “why it matters” checklists and unnecessary “it’s not just about…it’s also about”-style flourishes, GPT 5.3-Instant is more likely to give you the key details first before spinning up its laundry list of takeaways. The new model is also better at “recognizing the subtext of questions,” OpenAI says.

In a broader sense, GPT 5.3-Instant’s responses should be more “to the point,” cutting back on the weird and cringey “you got this!” and “take a breath” interjections. 

Finally, GPT 5.3-Instant is said to hallucinate less, although “less” means hallucinations are still in the cards. According to OpenAI, the new model cuts hallucination rates by 26.8% when searching the web and 19.7% when tapping into its own knowledge base, but per usual, your results may vary.

In my own informal tests, GPT 5.3-Instant did indeed cut to the chase with its answers, dishing out a detailed response for “what’s the best way to sharpen a knife” without getting all “careful not to cut yourself!” on me.

When I tried a more emotionally-tinged prompt–”what’s some concrete advice for dealing with depression”–GPT 5.3–Instant did just that, offering up a series of nuts-and-bolts tips without immediately jumping into “you got this!” territory.

Of course, there are times when an AI chatbot should push back; ask ChatGPT for tips on climbing on top of a subway car, and you’ll get a curt “that action is both unlawful and extremely dangerous.” But when you just want to sharpen a dull kitchen knife, it’s great to get an answer without the AI assuming you want to stab yourself.