UC San Diego researchers found GPT-4.5 was judged human 73% of the time in live conversations
Pexels
AI can pass the Turing Test in live chats, and the latest result lands with a chill. In a UC San Diego study, GPT-4.5 outperformed real participants at convincing judges there was a person on the other side.
The setup was harder to shrug off than a standard benchmark. Judges reacted to real-time exchanges rather than static prompts, then made a fast call based on conversation alone.
The unsettling part is how familiar the skill looks. The model didn’t need a body, a voice, or a biography. It only needed to sound like someone.
How did AI beat the human test
The study used a three-party version of the test. Judges chatted with both a person and an AI model, then chose which one they thought was real.
Four example games from two populations: (A) Prolific, (B) Undergraduate, (C) Undergraduate, and (D) Prolific. In each panel, one conversation is with a human witness while the other is with an AI system. C.R. Jones and B.K. Bergen / Proceedings of the National Academy of SciencesGPT-4.5 was identified as human 73% of the time when it was given a persona prompt. LLaMa-3.1-405B also crossed a striking line, getting picked as human 56% of the time with a persona prompt.
Those numbers give the finding its bite. The model didn’t merely avoid detection, it gave judges enough social cues to read it as the person in the chat.
Why does this test still matter
The Turing Test is a decades-old way to ask whether a machine can imitate human conversation well enough to fool a person. In the classic version, an evaluator chats without seeing the participants, then tries to tell the human apart from the machine.
It has always been more cultural symbol than clean measurement. Still, it remains the test people recognize when they want to know whether software can pass for one of us.
Alan Turing Yiming Ma / UnslpashThat makes the new result feel sharper. A chatbot doesn’t need consciousness, emotion, or self-awareness to create the impression that a real person is typing back. It only needs to be believable in the moment.
The risk shows up in ordinary places. Customer support, dating apps, social platforms, education, and political messaging all rely on quick judgments about identity, intent, and authenticity.
What should we watch next
The study stops well short of saying chatbots understand people. Its more practical finding is that some models can now perform personhood extremely well in short exchanges.
Clearer disclosure should become the next pressure point. When a bot can blend into casual conversation, users need stronger signals that they’re dealing with software, especially in places where persuasion or emotional vulnerability shapes the exchange.
The next fight is over labeling in chats where people make fast decisions about trust.
Paulo Vargas is an English major turned reporter turned technical writer, with a career that has always circled back to…
AI may have just won a literary prize. My heart weeps seeing it poison our love for books.
I had a hard time processing this news. As someone who has been deeply in love with stories since childhood and who grew up on the works of Arthur Conan Doyle, Terry Pratchett, J.R.R. Tolkien, and other such venerable authors, seeing an AI-written story win a prestigious writing award is hard to digest.
If you are unaware, the winners for the Commonwealth Short Story Prize for 2026 were announced, and three of the five winning regional stories have been found to be entirely or partially written by AI. Or at least that seems to be the consensus among readers. As a reader and an amateur fiction writer, this hurt me deeper than any other tale of AI corroding our lives.
Canva and Adobe are coming to Gemini, and they want to make everything chatty
Adobe and Canva are plugging into Google’s assistant, betting that creative work starts with a prompt, not an app icon
Canva and Adobe are moving deeper into Google Gemini, giving the assistant a bigger role before users ever open a design app.
Adobe says its "Adobe for creativity" connector is coming to Gemini in the coming weeks, giving users a way to describe tasks and send them through Adobe tools for imaging, design, and video. Canva is already rolling out its Connected App for Gemini in select English-language markets, with full availability coming soon.
Google wants Gemini to help build the next big scientific breakthrough
Gemini for Science pushes agentic AI deeper into real research workflows
Google is building Gemini deeper into the research workflow, starting with ideas, tests, and scientific literature.
At Google I/O 2026, the company announced Gemini for Science, an experimental suite built around agentic AI science. It targets the manual work behind discovery, including hypothesis building, computational testing, and literature review.
