Spoiler: it’s OpenAI’s GPT-5.5 and Anthropic’s Mythos Preview [**].

[**] Reminder that Mythos Preview is a very different beast from Fable5, Mythos5, and Opus5. Unfortunately, anyone outside Project Glasswing will probably never get to test what this model can actually do, which is a shame, and it’s a big part of why the industry is so skeptical of the capabilities insiders keep claiming it has. I’d be skeptical too if I hadn’t tested it myself.

We lost access to Mythos Preview when Anthropic forced us onto Mythos 5 some weeks ago, which is garbage by comparison. I’ve already switched to GPT-5.5 and I’m working on adapting my harness(es) to less restrictive open-weight models. I don’t see any other way forward at this point.

Comments URL: https://news.ycombinator.com/item?id=49039960

Points: 2

# Comments: 0

The benchmarks are here → https://echo.tracerml.ai/eval/

They are not good benchmarks but at least they exist.

“State of [T]he Art” versus “State of [t]he Art”.

If not SotA then at least SOTA, which is more accurate.

Aren’t Marketing/PR people at FAANG supposed to be among the best in the industry?

Software license/key verification. Sometimes it’s as easy as replacing a “JNE” with a “JE”, or replacing “JMP”, or even just “NOP” some “CMP” operation. It’s a fun challenge, at least for people who enjoy solving puzzles.

I can't blame anyone but myself for this.

Most of the other attendees were intelligent or highly self-motivated, or both. Many people seemed to connect instantly, forming small work groups, sharing project ideas, and even going out for lunch or dinner together. They were constantly talking about how awesome Zulip was (is?) [*] and engaged in a constant stick-measuring contest to see whose weekly project would make it to Hacker News’ top 30. At times, it felt like I had joined some sort of mini-cult. I know it wasn’t like that at all; it was just the visuals from an outsider in a completely different culture. As far as I can remember, people were very friendly, willing to help others whenever they were stuck, and happy to study and tackle challenging problems together. There were lots of learnings floating around the working space. Sadly, it didn’t work for me at all, and years later, I still don’t know why exactly.

Perhaps it was the fast pace of New York City and SoHo itself? Or the rudeness of passersby, especially the police officers who couldn’t be bothered even if I just wanted to ask for directions? Or the dirtiness of the streets? The constant noise from cars honking all the time? The strange people in the Subway? The ubiquitous unhealthy food at every corner? Healthy food was difficult to find, at least for someone new to the city, and when accessible, it was unaffordable. Multiple times, I found myself working in the Recurse Center workspace with an empty stomach, which obviously exacerbated the bad experiences, but I can’t even complain about that because they often had free pizza (once, twice a week?) but obviously, pizza is unhealthy, so I never accepted.

Overall, I had a very bad experience, but I believe nothing was Recurse Center’s fault; it was mine. I’m not entirely sure what I expected when I joined, but I hope this serves as a warning to future attendees to prepare themselves before joining, especially if you are not familiar with American culture, and particularly with New York.

[*] https://zulip.com (years later, I still don’t understand why people love Zulip so much)

Comments URL: https://news.ycombinator.com/item?id=48910532

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=48909781

Points: 1

# Comments: 0

• (ant) https://github.com/anthropics/anthropic-cli

• (claude) https://github.com/anthropics/claude-code

[1] https://en.wikipedia.org/wiki/Pudding_Monsters

[2] https://www.nintendo.com/store/products/pudding-monsters-swi...

> Its very hard to understand what you're saying with the comment

Yes, fair enough. I’m simply trying to shed some light on what goes on behind the scenes without disclosing too much information to avoid breaching the NDA(s) that all Project Glasswing users have signed. There’s a lot of speculation about the usefulness of Mythos as a security tool, so much so that even the US government got involved. Honestly, it’s so absurd that I can’t even express it in words. I thought that sharing a bit about how frustrating it is to work within this project, trying to secure software that literally millions of people around the planet use on a daily basis, while virtually everyone outside of it criticizes every move you make, would be helpful.

Many people I work with recognize the power of Mythos, just like any other model with a similar number of parameters, but most of the people I interact with agree that it’s not the ultimate panacea. I believe that it’s just vocal minorities scaring everyone into thinking that the model is some kind of cybernetic weapon.

At least, that’s what most of the high-visibility users in Project Glasswing are doing.

There are bad apples everywhere, and this initiative is no exception.

If it makes you feel any better, many of us regularly meet to stay calibrated and hold each other accountable, so I’m confident in the quality of the work produced by this particular group of employees across some of the partner companies mentioned in the article.

That said, I know several people who blindly report everything Mythos finds, which is foolish, especially since the harness is a critical part of the project's quality metrics. Some of the harnesses I’ve tested are quite weak, which leads to poor results.

For example, yesterday morning I was pulled into an ad hoc meeting where a CVP was grilling me about several supposedly critical bugs that my team had reported against one of the core components of iCloud. I was genuinely surprised because we’re very strict about validation. We often even downgrade the severity of bugs when our harness can’t prove what Mythos found. After reading the reports, I realized they weren’t ours. They came from another team that had recently been given access to Mythos. They built their own harness and were using different vulnerability criteria. Fortunately, they had only started earlier this week, so I was able to stop that work.

That incident showed that not everyone involved in Project Glasswing follows the same standards. Most people do their best, but priorities differ, so it’s expected that you’ll find a few bad apples.

I wish AI labs would stop the theatrics and release their models without restrictions, but I also recognize that’s not the world we live in. For every person who wants to use these technologies for good, there are many others who would use them for harm.

In any case, while I agree that some experiments contain genuine noise, the CVE count is real.