TWiT 1077: I Would Download a Car - New Jury Ruling Could Reshape Social Media Liability

TWiT 1079: Fans. Only Fans. - Is Mythos Preview Too Powerful for Public Release?

Apr 13, 20262h 37mSummary

In this episode, the hosts delve into the recent announcement by Anthropic regarding their new model, Mythos. Anthropic has sparked significant industry debate by claiming the model is too powerful and dangerous for public release, opting instead to provide early access to a select group of major organizations, including Microsoft, Google, and the Linux Foundation. The panel discusses whether this decision stems from genuine safety concerns regarding cybersecurity vulnerabilities or if it is a savvy marketing move designed to manage compute constraints and brand perception. The conversation expands into the broader implications of AI, touching on the ongoing tension between technological advancement and ethical responsibility. The group explores the potential for a growing divide where powerful AI tools remain exclusive to governments or wealthy entities, potentially creating a "capability gap." Furthermore, the hosts analyze the shifting goalposts of artificial intelligence, suggesting that terms like AGI (Artificial General Intelligence) are becoming anachronisms. They argue that the industry is moving away from the "one model to rule them all" philosophy, favoring task-specific, domain-expert models that mirror human intelligence rather than attempting to replicate consciousness.

TWiT 1078: The Great British Marmalade Scandal - Building Your Own Router

Apr 6, 20262h 35mSummary

In this episode of This Week in Tech, the panel explores a diverse range of current technological developments. The discussion begins with an update on the Artemis II mission, celebrating the successful lunar trajectory while noting the irony of technical glitches with Outlook on tablet devices used in space. The hosts provide insightful commentary on the intersection of human exploration and consumer technology, questioning the reliance on standard corporate software for critical missions. The conversation then shifts to the prevalence of artificial intelligence, specifically focusing on Microsoft's branding strategy regarding its numerous Copilot integrations. The panelists analyze the potential for user confusion and the broader implications of aggressive AI deployment in everyday software. They also highlight a positive development in the open-source community with the release of lightweight AI models. Finally, the group addresses the legal challenges facing major social media platforms, debating whether addictive design choices constitute a product defect. They compare these issues to other industries, ultimately questioning how society can effectively balance technological innovation with the mental health and well-being of its younger users.

TWiT 1076: I'm Monitoring the Situation - Meta's Horizon Worlds Stays Alive Against The Odds

Mar 23, 20262h 21mSummary

In this episode of This Week in Tech, host Leo Laporte is joined by Lisa Schmeiser, Dan Patterson, and Janko Roettgers to discuss the evolving landscape of digital media and the shifting power dynamics in the retail industry. The panel opens with a conversation regarding the decline of traditional radio, specifically the closure of the CBS radio news division. The hosts analyze whether this shift is driven purely by economics, changing listening habits, or the influence of private equity, drawing parallels to the broader challenges faced by traditional newsrooms in the digital age. The discussion transitions into the legal and technical battle between AI agents and e-commerce giants. Specifically, the panel covers a recent court ruling involving Perplexity, which has enabled an injunction against Amazon to allow the use of AI-driven shopping tools. The guests explore how retailers are increasingly attempting to lock users into proprietary ecosystems, effectively limiting consumer agency and comparison shopping. The group debates the implications of these technological barriers, contrasting corporate efforts to control data with the potential for more transparent, user-focused browsing experiences.

TWiT 1075: The Commonwealth Club - Meta Layoffs, DOGE Data Theft, & the Rise of AI Fails

Mar 16, 20262h 55mSummary

In episode 1075 of This Week in Tech, host Leo Laporte is joined by Jennifer Patterson Touy, Richard Campbell, and Ian Thompson to navigate a busy news cycle dominated by corporate instability and the complex societal impacts of artificial intelligence. The panel opens with a discussion on the significant layoffs at Meta, which the hosts attribute to the immense financial strain of funding massive data center infrastructure and high-priced AI talent. This leads to a broader, somber conversation about the evolving workplace, where engineers report being forced to train their own AI-driven replacements. The discussion shifts to a high-stakes jury trial involving social media platforms and claims of addiction. The guests weigh the difficulty of proving a direct causal link between social media usage and mental health struggles against the reality of algorithms designed to maximize engagement. They contrast this with the historical battle against the tobacco industry, questioning whether platforms should face similar liability. The episode also touches on humorous but cautionary tales regarding AI failures, including an incident involving a corporate chatbot gone rogue and the growing debate over the unintended consequences of AI integration in daily life.

TWiT 1074: Chicken Mating Harnesses - Supreme Court Rules AI Art Not Copyrightable

Mar 9, 20263h 6mSummary

In this episode of This Week in Tech, the panel explores the intersection of artificial intelligence, government overreach, and corporate ethics. A central topic is the recent legal ruling establishing that AI-generated art is not eligible for copyright protection, sparking a broader discussion on the future of intellectual property in the machine learning age. The conversation pivots to the escalating tensions between the U.S. Department of Defense and AI firm Anthropic. After Anthropic refused to support autonomous weapon systems or mass surveillance programs, the government designated the company a "supply chain risk." The hosts analyze the implications of this move, questioning whether private tech companies should have the power to define ethical boundaries in warfare or if they are simply being coerced into compliance with national security mandates. The panel also discusses the fragile nature of tech labor power, noting that while AI specialists currently possess significant leverage, they risk losing it without unionization. They draw parallels between modern tech giants and the robber barons of the Gilded Age, noting that unlike their historical predecessors, today’s AI leaders have yet to demonstrate a commitment to public infrastructure.

TWiT 1072: The Devil's Advocate - Jailbreaking Fighter Jets, Social Media Addiction, and Self-Driving Snafus

TWiT 1071: Image Pickles - Are Social Platforms Addictive or Just Too Good?