davedx

I don't understand the article.

"I’d say this benchmark answers with a resounding, “Maybe.”

Mythos maybe really is better than the other current models at finding security bugs"

Yet in the results, I don't see Mythos?

It seems like a really well researched article with lots of results for other models, yet the title seems to be clickbait because the results don't contain Mythos, do they?

66k

Ryu vs Chun-Li full super parry win: https://x.com/LostMemeArchive/status/2067292940838535561

> A Japanese esports player and author who competes professionally at fighting video games. He specializes in 2D arcade fighting games, mainly those released by Capcom. Daigo is one of the world's most famous Street Fighter players and is often considered its greatest. His longevity is seen as rare in the world of competitive video games.

There's been a lot of discussion on the Interwebs about whether SpaceX's AI satellites are a) viable in an engineering sense, and b) economical. This project does not attempt to answer b) at all, but does try to have a basic look at a): given the FCC filing of where the satellites will be orbiting, what kind of impact will the cosmic environment there have on the specific compute tasks we expect them to be performing (namely, transformer inference)?

Full disclosure: the project was my idea, and I wrote the basic initial design draft, but Claude did all the rest. I definitely learned some things though. Feedback would be very welcome, in particular ideas about what the "software architecture mitigations" SpaceX might try.

But that sounds like the same workflow as Codex or Claude, except Cursor is only a harness without its own model? (Or do they have their own model?)

Have you heard of ASML? NXP?

Ignorant comment

It doesn't tell you it's doing that. Wait, now it does. Or does it?

The policy decision to silently nerf AI/ML code produced by Fable surely wasn't just something "accidentally imported via a bad hire"? It seems to me like Anthropic wants to control who can develop frontier AI models. Maybe from inside Anthropic that seems like a noble mission, but from the outside it seems like downright shady anti-competitive behavior.

I saw in latest news the decision has been partially reversed -- because of external pressure...

Read the examples Anthropic gave in the model card. They refer to extremely broad technology used across AI and ML.

Could this be legally construed as anti-competitive behavior?

Edit: I asked Claude. It replied:

> Consumer protection / deceptive practices. In the EU this would be a clear UCPD (Unfair Commercial Practices Directive) issue and potentially a DSA violation. In the US, FTC Act §5 prohibits "unfair or deceptive acts." Selling a product that secretly performs worse than advertised for a commercially self-serving reason, without disclosure, is textbook deception. The Samsung/Apple battery throttling cases are instructive here: Apple faced regulatory action across multiple jurisdictions specifically because users weren't told.

> Competition law. This is where "anti-competitive" gets complicated. Refusing to help competitors build competing products via your ToS is generally legal — you can decide who you license to. But covertly sabotaging output quality for a class of users while charging them full price crosses into different territory. Under EU competition law (Article 102 TFEU), if a company with dominant market position uses covert technical means to disadvantage competitors, that's closer to abusive conduct than a legitimate ToS restriction.

I made a timesheet entry, invoicing, and basic bookkeeping system for my freelance business. It works pretty well, I used "spec driven development" with Codex and it one-shot the entire application except for the PDF invoice layout which needed iteration.

Don't count on it. They might not break out inference from training.

Agree

https://en.wikipedia.org/wiki/I_Have_No_Mouth,_and_I_Must_Sc...

Virtue-signalling or just the daddy?

https://www.berkshirehathaway.com/

I don't know if I agree that's the only sane workflow; the problem is, I am way less invested doing code reviews of agents than I am reviewing code by human colleagues.

I would love to be able to say I pay the same amount of attention and am just as diligent and communicate as clearly with an agent, but it wouldn't be honest: I scan agent PRs for obvious mistakes or misinterpretation of what they've implemented.

With human colleagues I usually know them and their style, their way of working, so have a better idea what to look for. You also have a genuine return on providing feedback that helps coworkers learn and improve, whereas with agents, all the feedback you write is gone when the thing gets merged (unless your org has some kind of shared memory for its agents).

I don't have the answer for what the future looks like, but I suspect agent-type-1 reviews agent-type-2 is actually where we'll end up.

Ha. I've worked in a fairly strongly regulated sector (energy, in the Netherlands), where I collaborated closely with our head of compliance, and she heavily over-interpreted the regulations while I often tried to find more pragmatic solutions.

I think adherence to regulation and compliance is nothing to do with whether you're a SWE, a risk officer, or C-level, and everything to do with your own principles, ethics, professional attitude, and pragmatism.

This could be super interesting to watch as the IPO gets closer.

Someone, somewhere, is doing a lucrative arb trade. And someone, somewhere, will be caught on the wrong side of it.

"Current research" Citation needed. Multiple, given the extraordinary claim.

Start gradually converting your equity to bonds is the standard advice on that timeframe. If you're dreading equity drawdowns, that's what fixed income is for.

Eh, Tesla had a relatively normal growth company valuation for a while when they were growing strongly. The problem is the stock still hasn't compressed the multiple back down as growth stagnated... because the market swapped out "valuation based growth" for "call option on robotaxi success" at the blink of an eye.

Your job URL gives a 404. Sounds awesome though, good luck!

- cheaper - much less upfront capex, lower operating costs

- removes nasty geopolitical dependencies on eg gulf state oil and gas; costs are more predictable

- easier to plan and build because the base units are much smaller

Variability remains a valid objection, to which the main answer today is "batteries. Lots of batteries. And other cheaper longer duration storage, like sand heat storage, vanadium flow, and good old pumped hydro."

Yeah the relatively recent paper that takes LCOE and adds back a bunch of cherry picked system costs is a PITA to refute because it's inherently complicated and actually has some good points.

The problem is every good argument for renewables will always inevitably have someone come up with some kind of counterpoint that on the surface may seem reasonable to those without the time or inclination to deeply research it all.

Energy is complicated.

FWIW I agree with a sibling poster who said to just say "its cheaper".

Meanwhile, Google...

I recently went through this process with Microsoft for Office365 and it was reasonably well executed: it needed escalation and three separate callbacks to first verify, then reset my password, then reset my MFA (I changed my phone and lost the lot).

"until I can get a few shares at a reasonable valuation"

I doubt you will. Most likely IPO reference price will be like SpaceX's, 100x ARR or so.

You're better off buying Google who own a huge chunk of Anthropic at a much saner average.

SEEKING WORK | Remote | I'm in the Netherlands

Hi! I work for clients in the US and Europe on projects across energy, edtech and finance. My most recent work has been heavily AI oriented, so I now call myself an AI Engineer, but I'm also a full stack web developer and throw up infra with terraform in GCP or AWS.

I'm familiar with a bunch of technologies, including but not limited to TypeScript, python, Unity3D, WebGL, .Net core, and others. I've been in this industry for 20+ years now and what I've seen work is the same now as it was then: solid communication, simple architecture, pragmatism, and velocity.

My rate is $110/hour and you can reach me at davedx@gmail.com - let's get shippin'!

The recruiters I know haven't changed their methods that much. They understand AI and some of the implications but at the end of the day it's the end client who has a job spec and they still follow whatever is in that spec.

If you want to make money, buying SpaceX stock isn't the way to do it

This is about valuation not ESG