My bet is that Mythos is still over-hyped and the cybersecurity fear and guardrails are mostly marketing to force company partnerships through Glasswing and get public attention.

▲

Flere-Imsaho 28 minutes ago | parent | next [-]

The UK gov disagrees with you:

https://arstechnica.com/ai/2026/04/uk-govs-mythos-ai-tests-h...

https://www.aisi.gov.uk/blog/our-evaluation-of-claude-mythos...

▲

miohtama 2 hours ago | parent | prev | next [-]

Mythos is from the same guy who did "GPT-2 is too dangerous to release"

https://naokishibuya.github.io/blog/2022-12-30-gpt-2-2019/

▲

oceansky 2 hours ago | parent | next [-]

He was kinda right.

Lawyers, doctors, students, teachers. Lots of people using GPT models carelessly in harmful ways.

	▲	alasano an hour ago \| parent [-]
		Obviously not what he meant at the time but hilarious(ly sad) in retrospect.

▲

uselessTA 12 minutes ago | parent | prev | next [-]

The claim I remember was that releasing it would start an arms race for AGI, which was absolutely true

▲

killerstorm 44 minutes ago | parent | prev | next [-]

"Malicious use" means spam, propaganda bots, etc. It's nice to give people who work on spam filters some heads-up.

▲

InsideOutSanta an hour ago | parent | prev [-]

People quote the "GPT-2 is too dangerous to release" thing as if it were wrong, but given all the slop all over social media and how it's used to create division and attack social cohesion, he was clearly right.

▲

bel8 2 hours ago | parent | prev | next [-]

It worked for OpenAI when GPT 3 was deemed too dangerous to be released. This is just a spin of that.

▲

hootz 2 hours ago | parent [-]

I still remember it. "Open"AI going API-only because GPT-3 is really really dangerous, so forget the Open in our name and all of that, you can't download our models anymore and must request access to them because they pose a THREAT.

Fast forward to today and GPT-3 has laughable performance.

▲

shoeb00m an hour ago | parent [-]

Even back then there were plenty of people who got fooled by AI generated articles. It's easier to spot AI writing now because we are so used to it. They were right to be concerned; not that it achieved much since oss models run laps around gpt-3 now.

	▲	hootz an hour ago \| parent [-]
		But it seems like that was not genuine concern, but instead a tactic to pivot to closed models and an API service with an excuse to do so, breaking the public's expectation that they would be a non-profit making open models, like their name implies.

▲

geerlingguy 2 hours ago | parent | prev | next [-]

Bingo.

"We had to do extra work to make this safe because it's so advanced and dangerous..." how many times can they trot out that line before it loses its effect entirely?

▲

copperx an hour ago | parent | next [-]

Only three times, if fables are right.

	▲	TaupeRanger 40 minutes ago \| parent [-]
		The Startup Who Cried Unsafe, by AIsop

▲

aesthesia an hour ago | parent | prev | next [-]

I mean, they do actually describe what that extra work was, and people elsewhere in this thread are complaining about the effects of those safeguards. So it's not like this is purely empty rhetoric.

	▲	zem 36 minutes ago \| parent [-]
		people are not questioning whether they did the work, they are questioning whether the work was really necessary (i.e. if mythos is really so good that it needs safeguards to prevent malicious actors from using it)

▲

OtomotO an hour ago | parent | prev [-]

With homo "sapiens" "sapiens"? A few decades at least.

▲

CSSer 42 minutes ago | parent | prev | next [-]

Yes, and "in collaboration with the U.S. Government" feels like a very gross ploy at appeal to authority. You don't need Mythos or really any SotA frontier model to make malware or do extensive penetration testing/reconnaissance already. Sure, Mythos might be faster/more efficient, but the cat has been out of the bag for awhile. Even the terminology "infrastructure providers" practically screams "Enterprise leads".

▲

whazor 19 minutes ago | parent | prev | next [-]

I think all models can find vulnerabilities if read the entire code base. Or intelligently combine parts of the codebase. Especially with test loops.

▲

teaearlgraycold 43 minutes ago | parent | prev | next [-]

I know a security researcher at Google with access to Mythos. He says it's the "real deal" and that "there are career plans I had that are no longer viable".

▲

ls612 2 hours ago | parent | prev [-]

And to ensure that only USG-approved entities are allowed to secure their code.