cedws

11 ms·

241.
by cedws 6mo ago
There’s the argument that Anthropic has built Claude Code to use the models efficiently, which the subscription pricing is based on. Maybe there’s some truth to that, but then why haven’t OpenAI made the same move? I believe the main reason
242.
by cedws 6mo ago
Didn’t they also use Mythos to scan Linux many times over and it only found one DoS bug or something? I find it hard to believe there is only one security bug lurking.
243.
by cedws 6mo ago
Yes. And people will find a way to game it like they always do.
244.
by cedws 6mo ago
RDW tested it themselves. That isn't real world usage. Real world drivers will not pay the same attention, and they'll do stupid shit like game the driver attention mechanism so they can watch TikTok.
245.
by cedws 6mo ago
I know it’s not fair to attribute someone else’s actions to Altman, but his words about upholding democracy feel a bit hollow given his relationship with Brockman. Brockman gave a $25 million donation to a Trump super PAC. As a reminder, Tr
246.
by cedws 6mo ago
“Supervised self-driving” feels like an oxymoron. It’s either safe enough to be completely autonomous or it’s not. There’s no way a human driver being assisted by self-driving is going to be ready to take over at a moments notice with the s
247.
by cedws 6mo ago
Is this a bug, security vulnerability, or just an oversight? It’s not clear to me. As a precaution would it be a good idea to run that reset command for all apps?
248.
by cedws 6mo ago
Fuck the bureaucrats responsible for this. I’m so sick of being completely powerless to fight any of this, being forced to sit and watch. Writing to my MP changes nothing. Signing petitions does nothing. The Government doesn’t give a fuck.
249.
by cedws 6mo ago
In my experience asking OpenAI or Anthropic models to do anything FAANG doesn’t want you to do is usually rejected. For example reverse engineering an app, cracking your own device, etc…
250.
by cedws 6mo ago
I feel like a bit of an idiot because I didn’t know this either. I just assumed OR was another startup burning money to provide models at cost. OpenRouter is a valuable service but I’ll probably try to run my own router going forward.
251.
by cedws 6mo ago
I saw this coming. Anthropic wants to shift developers on to their platform where they’re in control. The fight for harness control has been terribly inconvenient for them. To score a big IPO they need to be a platform, not just a token pip
252.
by cedws 6mo ago
Yes they use Slack. Mo seems like a nice product but I think it would be difficult to push for, we try to minimise the number of software vendors we rely on so we prefer to run small things like this ourselves.
253.
by cedws 6mo ago
That fab will never be delivered. In five years you might see the manufacturing equivalent of a person dancing in spandex.
254.
by cedws 6mo ago
I was looking for something kind of like this. We wanted to gate release PRs on approval from people in the company that don’t use GitHub.
255.
by cedws 6mo ago
>Iran didn’t escalate against anyone except their aggressors What about the missiles launched at Dubai?
256.
by cedws 6mo ago
It's not a matter of knowledge, it's a matter of resources. It takes billions of dollars of hardware to train a SOTA LLM and it's increasing all the time. You cannot possibly hope to compete as an independent or small startup
257.
by cedws 6mo ago
I understand that, but Anthropic is doing nothing to throw those grassroots researchers a lifejacket. This is the beginning of the end for independents, if it continues on this trajectory then Anthropic gets to decide who lives and who dies
258.
by cedws 6mo ago
Not only companies, they're going to be taking applications from individual researchers. No doubt that it will only be granted to only established researchers, effectively locking out graduates and those early in their career. This is
259.
by cedws 6mo ago
In practice this doesn't work though, the Mastercard-Visa duopoly is an example, two competing forces doesn't create aggressive enough competition to benefit the consumer. The only hope we have is the Chinese models, but it will a
260.
by cedws 6mo ago
More than killer AI I'm afraid of Anthropic/OpenAI going into full rent-seeking mode so that everyone working in tech is forced to fork out loads of money just to stay competitive on the market. These companies can also choose to
261.
by cedws 6mo ago
I want to experiment more with agents but my employer only pays for Claude Code, and TOS disallows using the subscription API for other purposes. Anyone else in the same boat? Token based pricing also gets expensive fast.
262.
by cedws 6mo ago
Sounds like a snake pit. None of them can be trusted. If we have to rely on companies to self appoint a benevolent ‘AI dictator’ we’re fucked. The only high profile person in AI I’d consider perhaps worthy of trust is Demis Hassabis.
263.
by cedws 6mo ago
Nah, there's a lot of truth to it. Phone snatching and looting is common but underreported because nothing ever happens. I've travelled quite a lot and other than Brussels, London (my home city) is the only one where I'm cowe
264.
by cedws 6mo ago
With Claude Code the problem of changes outside of your view is twofold: you don't have any insight into how the model is being ran behind the scenes, nor do you get to control the harness. Your best hope is to downgrade CC to a versio
265.
by cedws 6mo ago
It's objectively worse on the Ukrainian side. Imagine you haven't been able to leave your house in 4 years for fear you'll be grabbed by a draft officer. Russians do not know this fear. To boot, many Russian men have been pai
266.
by cedws 6mo ago
Hard to feel the same sympathy for Russian men to be honest, I've seen many gallivanting abroad, whilst majority of Ukrainian men are stuck either in hiding in their own country or have been sent to the front lines. Only a few got out
267.
by cedws 6mo ago
Yeah, I gained a habit of constantly checking over my shoulders because of the people who will speed past you on e-bikes with very little room. Even parents with their kid in the back ride like mad.
268.
by cedws 6mo ago
No but I have now. It’s hard to tell from that few seconds but it doesn’t look like it’s really putting the developer in the driving seat, just providing a minimal escape hatch for manual edits.
269.
by cedws 6mo ago
Yeah, this model where you don't get an editor anymore feels like a step backwards. I don't want to give up LSPs, being able to step into/rename functions and stuff like that. I should still be the one in control of the code
270.
by cedws 6mo ago
I quite like it, it’s a perfect signal for me to stop reading.

More ›