In the last weeks I experimented with Claude Computer Use to automate some daily tasks (via its Ui.Vision chat integration, see https://forum.ui.vision/t/v9-5-0-brings-computer-use-ai-chat... ) - and the results are mixed. Claude gets things wrong way too often to be useful.
Has anyone done any comparison Claude Computer Use vs OpenAI Operator? Is it signifcantly better?
100% this. I am using the open source Ui.vision to automate some business tasks. Works well, but only 10% of the work is for automating the main workflow, 90% of the work goes into error and edge case handling (e. g. Internet down, website (to scrape data from) down, some input data has typos or the wrong date format, etc).
A human can work around all these error cases once she encounters them. Current RPA tools like Uipath or ui.vision need explicit programming for every potential situation. And I see no indication that Claude is doing any better than this.
For starters, for visual automation to work reliably the OCR quality needs to improve further and be 100% reliable. Even in that very basic "AI" area, Claude, ChatGPT, Gemini are good, but not good enough yet.
Wow, what a click-baity misleading headline. Key point is that this number is going down:
"The big news is that more and more people are opting to trade their Teslas for an EV from a legacy automaker"
This might be me in a few years. Tesla stopped being innovative years ago. No surprise, its CEO is high on drugs most of the time and spends his days (and nights) posting childish memes.
I sort of wish car makers would stop being innovative. I do like the reliability and safety that modern cars provide but would like fewer bells and whistles. Affordability in both purchasing and maintaining is important to me. Just watching car commercial makes me think the industry is not prioritizing these things.
Your post is just as misleading. The ceo is irrelevant and his outward appearance speaks nothing to performance.
The more realistic assessment is that the path of innovation approaches a limit. Tesla was first to market, after that there’s not much left to innovate other than fusion powered cars.
Tesla created the market and they are losing first mover advantage as competitors move in and set up shop. The market is saturating and this is actually expected.
Totally disagree. Just one example: Tesla has only two global mass market models (3+Y). There is plenty of room to innovate on this topic alone. Model 2? Model whatever. Nothing new in years.
You mean planned obsolescence innovation with different but useless designs? They change the shape of the iPhone all the time just to keep it feeling new.
Tesla has been innovating on battery tech. Improving performance, efficiency and staying power instead of working on gimmicky features that are more for marketing.
What you said is all correct, but there are some positive steps as well, for example ELLIS: https://ellis.eu/
ELLIS is a great way to compete with the top US universities. Much more is needed, especially access to computing power, just as the CERN for AI proposal suggests.
> and debug if it errors or isn't what you wanted.
That is a potential problem Because at this point the user will still need good HTML/Selenium/coding knowledge to debug. We ran into the same issue using chatgpt generated scripts for ui.vision. Once a QA person has to figure out why some generated code does not work, it becomes a hassle and removes any initial advantage over the classic record & replay approach.
I agree, to really debug they need coding knowledge today.
But there's something in between where you can say "try again" or even give high level feedback like "don't click the element with that exact title, just click whichever one is first in the list". My working hypothesis is that these are bigger than the ones where you need coding knowledge. Certainly this category will only grow as language models become better.
What has a "German copyright cease and desist" to do with privacy, or the lack of privacy? And who is "them"? A few sources and details instead of just ranting would be useful.
If I recall correctly, some of these "cease and desist" cases involved tricking people into sharing specific copyrighted material, then asking the internet providers to identify users who violated copyright, and sending them cease an desist letters together with an invoice (which seems to be peculiarity of the German legal system).
If people had stronger privacy, these sorts of things wouldn't work.
Wow, Twitter is collapsing much faster than I expected. With PG and some other high-profile accounts gone, many will loose interest in their Twitter feed fast. Rinse and repeat.
I thought it'd collapse on the tech side before the policy side. Rather shocked the mask has come off this quickly on what "free speech" actually meant.
While I sort of expected the same... I think our HN crowd (myself included) is biased to assume the importance of technology more than the importance of the social dimension. But also, I think Musk put a heavy hand on the tiller far faster than I expected he would (the wise thing to do would have been to assume there was much to learn; he seems to have stomped into his new company with a belief he knows what's best, and that's not meshing well with what was already there).
But we should remember our own tech-first biases. Twitter ran in frequent-fail-whale mode for months with users accepting that because it fed their social needs. The moment it stopped serving those needs, people started leaving no matter how good the tech is.
Large distributed systems that have already been built can often limp along for a very long time before falling over. I would give Twitter at least another 3-6 months for stuff to start breaking.
Agreed entirely, but I thought the "free speech" stuff would last longer than that 3-6 months, if for no other reason than to avoid the embarassment of it.
On what level do you mean? 2fa already broke at one point (no clue if they fixed it I don't use two-factor as my twitter account is not terribly important)
You'll know the level when it happens - I am referring to multi-day outages. 2fa being broken doesn't surprise me at all. I bet the didn't want to pay the bill.
Oh yeah the full "can't access the site" could be a long way off. Could also be tomorrow if the wrong thing starts breaking and no one left understands how to maintain it.
No offense, but pg doesn't really post all that much stuff that would make me reconsider Twitter as a platform if he left. And, to be fair, neither does anyone else. Twitter is a marketing platform not a social network.
I use it primarily as a RSS feed and the occasional "get up to speed with the latest news fast" alternative.
Has anyone done any comparison Claude Computer Use vs OpenAI Operator? Is it signifcantly better?