🛒 Perplexity Releases AI Shopping Agent

Hey TEDchats Fam!

Happy Thursday! Remember when shopping online was supposed to save time? These days we spend hours scrolling, comparing prices, and second-guessing ourselves. Now, imagine outsourcing all this effort to a tireless AI assistant.

🧑‍💻 WEEKLY PROMPT

🍕 QUICK BITES

AI Agents Are Coming for Your Shopping Cart!

Perplexity has unveiled a cutting-edge AI shopping agent, Buy With Pro, that can navigate websites, select products, and even "checkout" for you. While still a work in progress—sometimes taking hours to complete a purchase—this innovation offers a glimpse into a future where AI manages your online shopping. Competitors like Google and OpenAI are reportedly hot on its heels, with AI agents poised to transform how we buy everything from everyday essentials to plane tickets. Amazon, too, has entered the fray with Rufus, a generative AI-powered conversational shopping assistant, underscoring the fierce competition to revolutionize retail (see Perplexity’s announcement).

Intel CEO Steps Down Amid Challenges

Pat Gelsinger, Intel's CEO, retires after a tumultuous tenure marked by bold moves and major hurdles. While his ambitious strategies, including massive U.S. chip factory investments, aimed to revive Intel’s edge, the company faced significant setbacks, including record losses and missed targets. What’s next for Intel? CFO David Zinsner and GM Michelle Johnston Holthaus step in as interim co-CEOs during the search for Gelsinger’s permanent successor (see Intel’s announcement).

From the NBA to High-Tech

Omri Casspi, former NBA athlete and Israel’s basketball trailblazer, has scored big in the venture capital game with the launch of Swish Ventures, a $60 million fund. The fund will focus exclusively on early-stage startups in AI, cybersecurity, and cloud infrastructure—narrowing the scope from his first venture, Sheva Capital. Casspi plans to invest $5–$7 million per deal in about 10 companies, spotlighting seasoned entrepreneurs poised to redefine their industries (read more).

🧰 AI TOOLBOX 

🤯 MIND SPARK

Amazon's Nova Models Are Here to Transform AI: Amazon has introduced Nova, a suite of foundation models for text, image, and video processing in Amazon Bedrock. It includes models like Micro, Lite, Pro, and Premier (Q1 2025), plus Canvas for image generation and Reel for video creation. Nova is reportedly fast, cost-effective, and supports tasks across 200 languages (see Amazon’s announcement).

ChatGPT Search's Fact-Check Flop: OpenAI’s ChatGPT search, a competitor to Google and Bing, struggles with accuracy—researchers found it confidently flubbed 146 of 200 quote sourcing attempts, showing uncertainty in just seven cases. The takeaway? Confidence doesn’t mean correctness (see analysis).

Censorship Concerns in Open Source AI: HuggingFace CEO Clement Delangue warns of the rise of Chinese open-source AI models, praised for their technical skill but criticized for censorship on topics. He cautions that dominance by one nation could spread its cultural values, affecting freedom of information and AI ethics (listen to the podcast [French]).

€11 Million Raised to Accelerate Precision Oncology: Orakl Oncology has raised €11 million to launch AI-powered solutions, O-Predict and O-Validate, aimed at transforming oncology drug development. Focused on colorectal and pancreatic cancers, Orakl plans to accelerate precision oncology and form strategic pharmaceutical partnerships to more precisely predict and validate drug responses (see Orakl’s announcement).

Clarifai Simplifies AI Compute Management: Clarifai has launched a new tool that lets businesses manage and optimize all their AI compute resources from one easy-to-use dashboard, offering the first compute orchestration capability for AI workloads across any AI model. This helps save money and makes scaling AI much simpler (see Clarifai’s announcement).

Meta’s Moderation Misstep and AI’s Unmet Fears: Meta’s president of global affairs, Nick Clegg, admits Meta’s moderation is overly aggressive, causing unnecessary takedowns. He acknowledges high "error rates" and pledges to better balance free expression with enforcement. Notably, Clegg adds that fears of generative AI driving election deepfakes and disinformation largely didn’t materialize (see Clegg’s full update).

Use AI as Your Personal Assistant

Ready to save precious time and let AI do the heavy lifting?

Save time and simplify your unique workflow with HubSpot’s highly anticipated AI Playbook—your guide to smarter processes and effortless productivity.

Download the free guide today.

🎨 AI ART

In partnership with