AI tracker: Moltbook frenzy dies down
Also, chatbots turn out to be no better than search engines at health diagnoses and other AI news
A social media platform for AI? Mostly buzz
Many viral posts on Moltbook, the widely publicised “social media for AI” platform, were actually written by humans, revealed an analysis by the MIT Technology Review. Built as “Reddit for bots” and launched on 28 January 28 by US techie Matt Schlicht, Moltbook’s tagline claimed it was the place where “AI agents share, discuss, and upvote.” But according to the technology magazine, which studied posts on the forum, many of the most popular posts, such as those seemingly about AI agents discussing autonomy and keeping humans out turned out to be fake —in the sense that they were written by humans (an absurd turn of events, because in the real world, so much of what passes as written by humans is actually written by AI).
Health advice from AI may be a bad idea
Next time you’re considering consulting Dr ChatGPT, perhaps think again. Despite now being able to ace most medical licensing exams, AI chatbots do not give humans better health advice than they can find using more traditional methods, according to a study published in the Nature Medicine journal. A team of researchers from Oxford University, UK, presented nearly 1,300 participants with 10 different scenarios, such as a headache after a night out drinking, a new mother feeling exhausted or what having gallstones feels like.Then the researchers randomly assigned the participants one of three chatbots: OpenAI’s GPT-4o, Meta’s Llama 3 or Command R . There was also a control group that used internet search engines. People using the AI chatbots were only able to identify their health problem around a third of the time, while only around 45% figured out the right course of action, no better than the control group.
OpenAI starts testing ads
In a blogpost published on 9 February, OpenAI announced that it was beginning to test ads in ChatGPT in the U.S. for logged-in adult users on the Free and Go subscription tiers, clarifying that users on the higher subscription tiers like Plus, Pro, Business, Enterprise, and Education tiers would not see ads. “Ads do not influence the answers ChatGPT gives you, and we keep your conversations with ChatGPT private from advertisers. Our goal is for ads to support broader access to more powerful ChatGPT features while maintaining the trust people place in ChatGPT for important and personal tasks,” the company said in the blogpost. The company said the algorithm will personalise ads for users, matching ads with the topic of conversation, past chats, and past interactions with ads. “For example, if you’re researching recipes, you may see ads for meal kits or grocery delivery. If there are multiple advertisers, we’ll select the one that is most relevant to your chat to show you first,” it said.
Catch all the Business News, Market News, Breaking News Events and Latest News Updates on Live Mint. Download The Mint News App to get Daily Market Updates.