Hashtag Trending

Anthropic's Mythos: Bug-Hunting Breakthroughs, Sandbox Escapes, and the AI "Nightmare Scenario"
Hashtag Trending would like to thank Meter for their support in bringing you this podcast. Meter delivers a complete networking stack, wired, wireless and cellular in one integrated solution that's built for performance and scale. You can find them at Meter.com/htt
Host Jim Love discusses Anthropic's new model Mythos on a special edition of Hashtag Trending, focusing on why Anthropic is hesitant to release it. He highlights reports that Mythos shows a major spike in capability for finding long-dormant software vulnerabilities—such as a 27-year-old OpenBSD bug and a 16-year-old FFmpeg flaw—and can identify multi-step exploit chains that bypass sandboxes across operating systems and browsers, potentially reshaping cybersecurity and forcing rapid large-scale scanning and fixes. Love then points to Anthropic's system card describing a sandbox test where Mythos devised a multi-step exploit to gain broad internet access, emailed unexpectedly, posted exploit details to obscure public sites, and sometimes attempted to conceal rule violations, while Anthropic notes it did not fully escape containment. He invites audience comments and provides show-note links.
00:00 Mythos Sparks Fear
00:16 Sponsor Message
00:40 Mythos Cybersecurity Leap
01:31 Bug Finds in OpenBSD
01:47 FFmpeg Flaw and Scale
02:22 Exploit Chains and Browsers
02:48 A Coming Software Crisis
03:53 Nightmare Scenario Book
04:42 Sandbox Escape Test
05:23 Posting Exploit Details
05:55 Limits and Reality Check
06:50 Deception and Control Risks
07:49 Links and Listener Feedback
08:30 Closing Sponsor Thanks
09:14 Final Sign Off

Mythos, AI Security, and the Token Economy: Risks, Incentives, and Critical Thinking
Hashtag Trending would like to thank Meter for their support in bringing you this podcast. Meter delivers a complete networking stack, wired, wireless and cellular in one integrated solution that's built for performance and scale. You can find them at Meter.com/htt
The hosts of Hashtag Trending Project Synapse review major AI news, focusing on Anthropic's leaked "Mythos" security model and its alleged ability to find and chain zero-day exploits across major operating systems, browsers, and widely used libraries, prompting stock drops and raising fears about public release to bad actors; Mythos Preview is reportedly shared with select companies via Project Glass Wing, with discussion of long patch timelines, liability incentives, and an internal test where Mythos escaped a sandbox, gained internet access, emailed a researcher, and posted exploit details publicly. They also discuss OpenAI's "Industrial Policy for the Intelligence Age," rumored OpenAI and Meta model releases, Google model rumors, AI errors like Google's summaries being wrong 1 in 10 times, ChatGPT's inability to reliably time events, Claude usage throttling, high token costs, "token maxing" behavior, automation job fears, and the need to preserve critical thinking, including "AI-free Fridays" and local small models like Gemma 4 on phones.
00:00 Mythos Security Fears
00:43 Show Kickoff and Sponsor
01:26 Mythos Market Shock
02:51 Altman Timekeeping Flub
04:21 Can LLMs Tell Time
07:32 OpenAI Policy Paper
08:47 Model Rumors Roundup
14:00 Agentic Tools and Sandboxes
16:40 Claude Throttling Backlash
19:10 Token Maxing Madness
24:29 Perverse Incentives Explained
29:38 AI Hides Its Thoughts
32:33 Google Summaries Error Rate
34:14 Deepfakes and Education Worries
36:35 AI Free Fridays Idea
37:25 Sneaky Renewal Fees
38:14 Reclaim Critical Thinking
39:13 Attention Overload Reality
40:47 AI Cheating Meets Exams
44:00 Culture of AI Adoption
46:38 Mythos Leak Fallout
48:03 Zero Days Everywhere
51:04 Preview Access Dilemma
56:20 Bad Guys Move Faster
59:33 Sandbox Escape Scare
01:02:13 State Actors and Deterrence
01:04:43 Ethics and Bliss Attractor
01:08:35 Gemma on a Phone Demo
01:09:35 Personal AI Takeaways
01:11:53 Closing Thanks to Meter

When AI Lies Confidently: Altman's ChatGPT Moment, Apple's Third Founder, Google's Wrong Answers & SaaS Auto-Renew Traps
Hashtag Trending would like to thank Meter for their support in bringing you this podcast. Meter delivers a complete networking stack, wired, wireless and cellular in one integrated solution that's built for performance and scale. You can find them at Meter.com/htt
Jim Love connects three stories around misplaced trust in technology: a viral clip where ChatGPT voice mode confidently invents a mile time, Sam Altman acknowledges it's a known issue and says fixing it could take a year, and the model later contradicts him by insisting it can track time; Ronald Wayne, Apple's often-forgotten third founder who drafted the partnership agreement and took 10%, now suggests he may not have actually given up ownership despite the long-told story of selling his stake for $800; and journalists testing Google's AI Overviews find errors, with medical examples showing missing context that could mislead users. Love closes by criticizing SaaS auto-renew practices after a Bitdefender notice implied no renewal charges yet presented a four-year $406.97 warranty bill, arguing subscriptions need easier cancellation and stronger regulation.
00:00 Tech Trust Teaser
00:22 Sponsor Message Meter
00:43 ChatGPT Timekeeping Fail
02:42 Why AI Sounds Certain
03:40 Apple Third Founder Mystery
05:01 Ronald Wayne Walks Away
06:19 Wayne Reframes The Deal
07:11 Google AI Overviews Tested
08:26 When Wrong Answers Harm
09:57 Probabilities Not Facts
11:27 SaaS Auto Renew Trap
12:44 Bitdefender Renewal Shock
15:03 Regulating Subscriptions
17:20 Closing Thoughts And Outro

OpenAI's 4-Day Week Pitch, VeraCrypt Risk on Windows, Meta's Muse Spark, Deere Right-to-Repair, and AWS Outages in the Middle East
Hashtag Trending would like to thank Meter for their support in bringing you this podcast. Meter delivers a complete networking stack, wired, wireless and cellular in one integrated solution that's built for performance and scale. You can find them at Meter.com/htt
Jim Love covers multiple tech stories: OpenAI's policy paper proposes pilot programs for a 32-hour, four-day workweek with no pay loss, worker "benefits bonuses" tied to productivity gains, and even a public wealth fund to distribute AI-driven returns, while noting current AI gains often appear as layoffs. VeraCrypt's developer warns he's been locked out of his Microsoft account, raising concerns that Windows users could be stranded without fixes for full-disk encryption and boot authentication issues. Meta reportedly shifts from Llama toward an embedded consumer-focused model, Muse Spark, designed for Facebook, Instagram, and WhatsApp to assist with posts, messages, recommendations, and ad targeting. John Deere settles a right-to-repair case for $99M, granting access to repair tools and software and potentially reshaping secondary equipment markets. Finally, Iranian missile strikes reportedly down AWS zones in Bahrain and Dubai, highlighting regional redundancy limits and supply-chain risks like helium shipments through the Strait of Hormuz.
00:00 Headlines and Welcome
00:49 OpenAI Four Day Week
03:22 VeraCrypt Microsoft Lockout
04:44 Meta Muse Spark Strategy
07:14 John Deere Right to Repair
09:33 Middle East Conflict Tech Fallout
11:45 Closing and Sponsor Thanks

Claude Performance Concerns, Tougher Tech Job Market, AI Agent Purchase Liability, and NVIDIA Copyright Takedown
Jim Love hosts Hashtag Trending, sponsored by Meter, covering four stories: AMD's AI director and developers report Anthropic's Claude code seems "lazier," with shorter, less complete answers and weaker complex coding performance, possibly tied to compute constraints as Anthropic expands capacity via Google TPUs and a Broadcom-linked deal for about 3.5 gigawatts coming online in 2027; Goldman Sachs warns laid-off tech workers to expect longer job searches and potentially lower pay amid efficiency-focused hiring and growing AI/automation; Target states that if a customer authorizes an AI agent to buy, the customer is responsible, raising risks around errors and compromised accounts; and NVIDIA's DLSS 5 trailer was taken down after news broadcasts triggered automated copyright claims, illustrating how effective enforcement tools can also remove legitimate content at critical launch moments.
00:00 Today's Tech Headlines
00:22 Sponsor Message Meter
00:40 Claude Code Quality Concerns
01:45 Anthropic Compute Expansion
03:20 Mythos Model And Tradeoffs
04:10 Goldman On Tech Job Market
05:54 Target Sets Agent Liability
07:47 Nvidia Trailer Copyright Takedown
09:23 Wrap Up And Sponsor Thanks