AI at Duolingo, Fact-Checking Your WhatsApp, and Shopping with ChatGPT
Last week, I mentioned the growing lack of support for misinformation research in the U.S. This week, Perplexity AI announced a new integration with WhatsApp, enabling users to fact-check viral forwards they receive from friends and family.
As per the news release:
You just forward any questionable WhatsApp message to Perplexity's number, +1 (833) 436-3285, and the AI will explain in seconds whether the claim holds up or belongs in the digital trash. You can forward screenshots, messages, images with a quote that Einstein definitely said, and any other bit of suspicious information.
I gave the service a try—while the responses are often impressive, I noticed that the references frequently point to sources like Reddit and Wikipedia, which raises some concerns. Give it a shot and let me know what you think.
OpenAI has rolled back a recent update to ChatGPT that users criticized for making the chatbot overly “annoying” and “sycophantic,” following widespread reports and screenshots showing it delivering excessive and exaggerated praise.
Duolingo is shifting toward a more AI-driven model and will phase out the use of contractors for tasks that can be managed by AI. Duolingo will also embed AI into its hiring practices, performance evaluations, daily workflows, and select strategic initiatives, aiming to fundamentally transform the way its employees operate.
OpenAI has announced that all users can now shop directly through ChatGPT Search, enabling them to browse products, compare options, and make purchases within the platform. The feature currently covers categories like fashion, beauty, home goods, and electronics, with plans to expand to more areas in the future.
According to OpenAI, these search results are not advertisements. The new functionality is available with the GPT-4o model and is being rolled out to ChatGPT Plus, Pro, and Free users.
