DeeperML

OpenAI's GPT-4.5 model has reportedly achieved a significant milestone by passing the Turing test, fooling humans 73% of the time into believing they were interacting with another person. This benchmark was established by researchers at UC San Diego, who conducted a three-way test where nearly 300 student participants had to distinguish between human and chatbot responses. The results revealed that GPT-4.5 not only convinced judges it was human more often than other AI programs like Meta's LLaMa and the older GPT-4o, but even surpassed the ability of actual humans to prove their own humanity. This achievement, which is awaiting peer review, marks a new era in AI's ability to mimic human conversation, raising complex questions about the nature of intelligence and human interaction.

The implications of GPT-4.5 passing the Turing test are far-reaching, challenging existing perspectives on AI capabilities and potentially reshaping our understanding of human intelligence. While some experts like AI scholar Melanie Mitchell argue that the Turing test is less about true intelligence and more about human assumptions, the fact remains that LLMs like GPT-4.5 are becoming increasingly adept at mimicking human imperfections in language and reasoning. This blurring of lines between human and machine could lead to various real-world applications, from advanced customer service bots to more immersive and personalized educational tools.

In addition to the advancements in AI capabilities, OpenAI is also making strategic moves in the education sector, offering free ChatGPT Plus subscriptions to college students in the US and Canada. This initiative, launched just before finals week, is a competitive response to Anthropic's "Claude for Education" and reflects a growing battle among tech giants for dominance in the $80 billion education AI market. By providing students with free access to GPT-4o, image generation, and advanced research tools, OpenAI aims to capture student loyalty early, potentially influencing enterprise adoption of their AI tools in the future. This strategic chess move highlights the increasing importance of the education market as a crucial battleground for AI companies.

ImgSrc: cdn.mos.cms.fut

References :

• eWEEK: OpenAI’s GPT-4.5 model has definitively beaten the Turing test, after it was found to be human 73% of the instances where it was prompted to adopt a human-like persona.
• venturebeat.com: OpenAI offers free ChatGPT Plus to college students just before finals week, escalating competition with Anthropic's Claude as tech giants battle for dominance in the $80 billion education AI market.
• www.techradar.com: ChatGPT Plus is now free for some students for two months, saving you $40.
• Analytics Vidhya: Imagine having a casual chat online, assuming you’re speaking to a real person. But what if it’s not? What if, behind the screen, it’s an AI model trained to sound human? In a recent 2025 study, researchers from UC San Diego found that large language models like GPT-4.5 could convincingly pass as human, sometimes more […] The post appeared first on .
• www.eweek.com: OpenAI’s GPT-4.5 model has definitively beaten the Turing test, after it was found to be human 73% of the instances where it was prompted to adopt a human-like persona. The Turing test, named after British computer scientist Alan Turing in 1950, measures a machine’s ability to exhibit human-like intelligence in conversation with a human evaluator. […] The post appeared first on .
• The Tech Basic: OpenAI and Anthropic Race to Win Over College Students With Free AI Tools
• THE DECODER: OpenAI expands ChatGPT's memory to include full conversation history.
• Search Engine Journal: This article covers ChatGPT expanding memory capabilities and now remembers past chats.

Classification:

• HashTags: #GPT4.5 #TuringTest #OpenAI
• Company: OpenAI
• Target: general public
• Product: GPT-4.5
• Feature: human-like conversation
• Type: AI
• Severity: Major