Navigate Select ESC Close

we have months left...

2026-04-09 Education
45.4k
1.6k
436
Wes Roth
Wes Roth
320.0k subscribers

Unlock all features

FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.

Description

Check out tastytrade here: https://tastytrade.com/unleashed Episode notes, links etc: https://natural20.beehiiv.com/p/the-mythos-problem-nobody-s-talking-about ______________________________________________ My Links 🔗 ➡️ Twitter: https://x.com/WesRoth ➡️ AI Newsletter: https://natural20.beehiiv.com/subscribe Want to work with me? Brand, sponsorship & business inquiries: [email protected] Check out my AI Podcast where me and Dylan interview AI experts: https://www.youtube.com/playlist?list=PLb1th0f6y4XSKLYenSVDUXFjSHsZTTfhk ______________________________________________ 00:00 testers are "freaking out" 04:47 tastytrade (sponsor) 06:40 ten trillion param models 08:46 Karpathy 13:00 AI alignment 16:15 can small models do the same thing? 21:15 digital hygiene #ai #openai #llm

Top Comments (10)

@soggybiscuit6098 2026-04-09

CIA freaking out that its backdoors are being uncovered 😂

193 14 replies
@1x93cm 2026-04-09

This is what happens when the NSA puts back doors in everything under the sun- even in the hardware. If theres a door. theres a way in.

112 10 replies
@chuyhable 2026-04-09

I find it very interesting that the leak news got buried. This whole thing is so interesting

28
@openroomxyz 2026-04-09

Well maybe this is just a very clever advertisment basically

17 4 replies
@isajoha9962 2026-04-09

I guess we have reached the birth of Skynet (Anthropic branch).

16
@the_flushjackson 2026-04-09

Here's a snippet from GKH (Linux kernel guy) about AI bug reports changing swiftly from slop to real (from an interview on The Register): "Months ago, we were getting what we called 'AI slop,' AI-generated security reports that were obviously wrong or low quality," he said. "It was kind of funny. It didn't really worry us." Of course, there are many Linux kernel maintainers, so for them, AI slop isn't as burdensome as it is for, say, Daniel Stenberg, founder and lead developer of cURL, where AI slop reports caused the cURL team to stop paying bug bounties. Things have changed, Kroah-Hartman said. "Something happened a month ago, and the world switched. Now we have real reports." It's not just Linux, he continued. "All open source projects have real reports that are made with AI, but they're good, and they're real." Security teams across major open source projects talk informally and frequently, he noted, and everyone is seeing the same shift. "All open source security teams are hitting this right now."

10 2 replies
@theomnisthour6400 2026-04-09

If you have a hitchhikers guide to the universe, it's exactly the time to panic and find a good exorcist

10 2 replies
@Bvoid-b 2026-04-09

If I were to hazard a guess, Anthropic is making a show of mythos because of the issue with the pentagon, trying to raise their perceived value to have more leverage. Its probably real, I think, but the show is for that.

6
@ryguy-c3f 2026-04-09

You can go study cybersecurity, but just remember, Mythos found vulnerabilities that expert cybersecurity architects couldn’t find for over 20 years. So you’ll never be ahead of Mythos.

4
@WesRoth 2026-04-09

Check out tastytrade here: https://tastytrade.com/unleashed

3 2 replies

Unlock the Data Inside
Turn Videos into Knowledge

  • Get FREE 10/day: transcripts, summaries, chats
  • Chat with videos, export text & PDF
  • $1 free API credit for RAG, chatbots & research

Free forever plan • All features unlocked

App screenshot