MIT vs Puck

MIT Professor Warns AI Can Create False Memories

A study by MIT Media Lab professor Pattie Maes, as reported by the Washington Post, demonstrates that AI can influence human memory. In an experiment, 200 participants viewed 24 photos; some saw AI-altered versions with changed facial expressions, added military personnel, or climate change cues. Those exposed to altered images had more false memories and were more confident in them. Maes calls this 'very dangerous' for legal proceedings and historiography. The findings were discussed at an AI conference in Malmö, Sweden. Another study showed that participants who interacted with a generative AI chatbot after watching a crime video had three times as many false memories. The research highlights the risk of AI blurring the line between human and machine thoughts.

• MIT Media Lab professor Pattie Maes conducted an experiment with 200 participants.
• Participants who saw AI-altered photos had more false memories and were more confident in them.
• The experiment involved altering facial expressions, adding military personnel, and climate change cues to photos.

MIT PhD Alex Zhang Discusses Recursive Language Models and AI Harnesses

In this episode of the Latent Space podcast, Swyx and Vibhu interview Alex Zhang, a PhD student at MIT known for his work on Recursive Language Models (RLMs), GPU kernels, and AI agent harnesses. Zhang discusses his involvement with GPU Mode and KernelBench, the concept of RLMs as a harness design where code is the primary tool, and the idea of harnesses as compositional generalizers that can improve model generalization across tasks. He talks about Prime Agent, an RLM harness built on Pi Mono, and his views on agent swarms, citing OpenAI's 10,000-agent experiment costing around $40 million. Zhang advocates for academics to take big research bets, explores alternative model architectures like Jev, and touches on open-ended research at Sakana AI, capability overhang, and the future of language models potentially being invisible swarms of agents.

• Alex Zhang is a PhD student at MIT discussing his research on Recursive Language Models (RLMs) and AI agent harnesses.
• Zhang is involved with GPU Mode and KernelBench, focusing on AI-generated GPU kernels.
• The podcast discusses Prime Agent, an RLM harness built on Pi Mono, and persistent subagents.

New formulation helps RNA vaccines withstand high temperatures

MIT engineers have found a way to stabilize the lipid nanoparticles used to deliver RNA vaccines, which could allow the vaccines to be more widely distributed.

Looper Insights launches AI chat to price CTV home-screen placements

Looper Insights has announced the launch of Looper AI, a conversational interface that allows studios, streamers, and their marketing, content, distribution, and partnership teams to query the company's CTV home-screen placement data in plain language. The AI-powered chat provides answers with source tables, enabling users to understand the value of their on-device placements, compare with competitors, and take data-backed numbers into platform negotiations. The tool addresses a market gap where 65% of streaming executives believe platforms charge premium rates without adequate proof, yet only 8% use third-party verification. Looper Insights also plans to offer public Model Context Protocol (MCP) access, allowing its data to be integrated into AI assistants like Claude, ChatGPT, and Copilot. This move positions Looper as a data source for AI-driven media planning and could increase pricing transparency in CTV advertising.

• Looper Insights launched Looper AI, a chat interface for natural language queries on CTV home-screen placement data.
• Only 8% of streaming executives use third-party verification for premium placement rates, while 65% believe platforms charge without adequate proof (State of Streaming survey).
• Looper plans to provide public Model Context Protocol (MCP) access for integration with AI assistants like Claude, ChatGPT, and Copilot.

Mike Judge to Reclaim Beavis and Butt-Head Rights

Mike Judge is expected to be able to reclaim ownership rights to Beavis and Butt-Head from Paramount in 2027, according to reporting based on Matt Belloni’s newsletter at Puck News. The characters were created by Judge in the early 1990s and have been controlled by MTV and later Paramount through a series of corporate transitions (including Viacom and Paramount Global). Paramount revived the franchise in 2022 with a feature film and new episodes on Paramount+, later moving additional seasons to Comedy Central. If rights revert in 2027, Judge would gain control over future productions, licensing and ancillary uses, while existing catalogs already licensed to Paramount-operated platforms would remain in place.

• Mike Judge may reclaim control of Beavis and Butt-Head from Paramount in 2027.
• The report cites Matt Belloni’s newsletter at Puck News as the source for the reclaiming timeline.
• Beavis and Butt-Head originated as Mike Judge shorts in 1992 and premiered as a series on March 8, 1993.

Gwyneth Paltrow reportedly to host Sam Altman dinner

Gwyneth Paltrow is reportedly inviting guests to a private, off-the-record dinner honoring Sam Altman at her Hamptons home on August 29, according to Puck. The brief article notes Paltrow's roles as an actress and founder of wellness brand Goop and her investment activity through Kinship Ventures, which she launched in 2021 alongside Moj Mahdara. Kinship has been reported as raising a fund in 2023 and is listed in SEC filings as an investor in OpenAI; it has also backed AI companies such as Lovable and the marketing platform Nectar Social. The piece references Paltrow's public commentary on AI and mentions Goop Kitchen signing a lease at Anthropic's San Francisco HQ.

• Gwyneth Paltrow is reportedly sending invites for a private dinner honoring Sam Altman on August 29 at her Hamptons home (per Puck).
• Paltrow is described as an actress and founder of the wellness brand Goop.
• Paltrow invests through Kinship Ventures, which she launched in 2021 alongside Moj Mahdara.