__natty__

Model card: https://storage.googleapis.com/deepmind-media/Model-Cards/Ge...

Model card: https://storage.googleapis.com/deepmind-media/Model-Cards/Ge...

OpenAI chosen not the sharpest tool in the shed

Most of us know what is it about. The problem is Sony and other companies have plenty of money for lobbying governments. What options do we have? I know it’s mostly technical forum but is there any lawyer or government representative who can help?

It's hard to believe Composer 2.5 is that good. I tried to compare it with GLM 5.2 or Opus 4.6 and it lacked thinking about the problem and critical reasoning. It's great for executing plans made by other models, but even then it does some weird code manipulation that is far from how other files around actually work.

Discussion about Leanstral 1: https://news.ycombinator.com/item?id=47404796

More and more I find myself trying to stop Opus from doing something stupid, and at every turn I need to tell it to stop overcomplicating things

Yeah, that’s my thoughts as well. I feel it’s great for benchmarks and some tasks while in other it tries to spend as much tokens as possible, tries to overcomplicate task and needs seconds or third round of steering that costs. With the scale Anthropic operates I bet it’s huge amount of extra money just to make sure their model works.

The only use case I see for this is for hotfixing while away from your computer. Other than that it seems irrational for me to use it.

Also audit trails. Good audit trail can save company (and you) in emergency as well. Useful for debugging and last resort of compliance data source.

That’s exactly what I was thinking. Feels like someone is playing a high-stakes game, putting on a show involving the US government.

Over the last few years, I’ve felt as though I’ve been living in a feverish dream all the time. Laws, regulations and general changes in the world are so detached from reality and so far removed from the reality they are meant to serve. And this is yet another example.

And we get the news the same time OpenAI releases 5.6. What a coincidence?

I’ve just been thinking about this and would like to ask those of you with more experience in the industry: do you prepare yourselves in any way for crises? Do you have any techniques for dealing with uncertainty?

Let’s say someone is entering the IT job market today. How should they prepare for the responsibilities of adulthood, starting a family, buying (perhaps on a mortgage?) their first flat, and planning for a family? I’m not writing solely in the context of AI layoffs but broader. Over the last 10 years that I’ve been in the IT industry from EU/US perspective, I’ve perceived the market as economic cycles, the sine wave of which has narrowed significantly (more often markets demands changes and instability) and accelerated in the past years.

The biggest win for me from this product is pushing developers to release on Linux.

I wonder why they didn’t introduce this yet. Same anbout other cloud providers. It’s not that this is some rocket science technology to achieve because they calculate near realtime usage anyway. Can this be because they earn more on people going beyond what they would set limit?

Because this is how LinkedIn “specialists” promotes LLM. The same specialists shouting about crypto a few years ago, then specialists about nft and now about how coding, architecture, accounting, law, medicine and basically every white collar job is solved and you just need enough money to pay for Opus/GPT.

And often it’s not perfect either. Just because one is true it doesn’t dismiss the other

Windows 11 ships with two versions of Outlook. There is Outlook Classic, the long-running Win32 desktop app built for power users, and there is the new Outlook, which Microsoft is pushing as the future of email on Windows. The newer one is built on WebView2 and is, in essence, a browser window that loads Outlook.com. If you have ever used both side by side, you already know which one feels faster and which one does not.

This is bizarre. Kyle Rubenok, according to his LinkedIn and GitHub @krubenok, is the senior manager for the outlook product. Isn't he taking any responsibility for how poorly his product performs? In many regular companies, such a manager would be fired for managing a product into decline. I guess I don't understand how big US corporations work.

As a Go developer I'm also really looking forward to their Go port of the filesystem watcher package

You plan to contribute to the TypeScript project?

I disagree. At least from my experience it still hallucinates a lot of times - it's just better at hiding that. That causes issues and it's hard to believe for me inexperienced person can spot this problems.

I disagree. Searching in google now IS using generative AI.

A year or two ago people were concerned that Google was losing its battle to OpenAI. Today Google puts heavy emphasis on AI search. Most common folks are now using them. I know in a minute or two many Kagi, DDG or other alternatives will reply, but these people were never a core part of Alphabet's user base. I'm an AI sceptic to some level, but it's hard to deny that "most" people are using AI (as an LLM or in other forms) to some extent today despite we like it or not.

I wonder how many people are scraping this thread right now and posting into llm something like “take the best ideas from this thread with highest chance of quick revenue”.

Anyway, I’m working on my manual skills of soldering.

Everything went off the rails after we invented rails.

Same thoughts. If I need more performant caddy alternative I'm going to use nginx at least it has some extras.

Truely good work! It’s responsive, clean and “onboarding” experience without signup walls is great. Good job.

I do not trust Anthropic anymore. They put in silent guardrails, reverted them later after people complained to save face, were loud and obnoxious about how their models are dangerous and should be regulated, and now this. Too much drama for a typical end-user. I'm sticking to alternatives even if they have a bit more smarter (for now) model than others.

Heh it got me then :)

Funny enough. I’m opening this on mobile internet connection and it stuck at loading spinner. I don’t know if the problem is with my internet (probably not) or support for mobile so I can’t even read the content.

Quote from the dashboard:

By using Fable 5 you understand that the model provider will retain agent request and output data associated with this model, regardless of your Cursor Privacy Mode setting.

This data will be used for automated and, where necessary, human harm prevention processes. Anthropic will never train on this data.