Closi

26 ms·

361.
by Closi 3y ago
Did they consider the UK market when naming this? Horizon is synonymous with a computing scandal involving buggy IT code.
362.
by Closi 3y ago
At the very least it's a bad look considering Nestle's history here.
363.
by Closi 3y ago
Disagree - both can complement each other in different circumstances. If I’m ironing and want to know when my first meeting is, chances are a VUI is better. If I want to see my whole days itinerary, a GUI is probably better.
364.
by Closi 3y ago
Why though? That doesn’t really answer the question.
365.
by Closi 3y ago
Yes - but why do it is the main point? Why not focus on doing one thing properly? Although you seem to obviously be attracting enough money so it's up to you how to spend/burn it :)
366.
by Closi 3y ago
Eh, a founder that effectively says 'we don't care if we give away the identity of our users if they are criminals' is not totally in line with my definition of an organization focused on privacy. At least not a definition of
367.
by Closi 3y ago
I mean they could focus on actually building out their own search engine as an example? (i.e. moving further away from using Google and Bing APIs) It's just a matter of focus with a team of that size.
368.
by Closi 3y ago
The person you are replying to is arguing the opposite - a future VUI could know your interests and just read you out the one relevant event rather than reading a list. Alternatively it could summarise into “there’s a standup comedy gig, a
369.
by Closi 3y ago
Just annoying when the attributes aren't in a logical order because they have been added to incrementally. I've got a table that contains the columns length, width, height and cube which is length x width x height) but cube is lik
370.
by Closi 3y ago
Add to the list: Ability to reorder columns :)
371.
by Closi 3y ago
I don't think number 1 is a differentiator anymore (there are so many boilerplate bootstrap templates that include OAuth). 2 is certainly a differentiator, but not mentioned anywhere in your post.
372.
by Closi 3y ago
The LLM can do the research itself, just current LLM's often don't by default (and require a bit of prompting). LLM's also use calculators (and again, if you prompt ChatGPT it will use a calculator to generate the result). Th
373.
by Closi 3y ago
> LLMs don't have any clue what the concept "research the question before answering" means. GPT-4 in ChatGPT will absolutely do research if you ask it (in the practical sense that if you say “research first” like in my ori
374.
by Closi 3y ago
GPT-4 absolutely can do more research. Adding this to a prompt will encourage GPT-4 to do web searches in the background to augment its answer.
375.
by Closi 3y ago
Treat an LLM as a confident smart person who isn’t an expert in anything, and doesn’t have access to resources to check their answer (unless you give it access). If you assume the above, it’s unsurprising that it doesn’t get your question r
376.
by Closi 3y ago
There is an off the shelf product for what is described in the article - it’s an ERP that includes a WMS (or even possibly just a WMS plus some ordering/forecasting software).
377.
by Closi 3y ago
Demo shows billing 2 minutes work as 0.2 hours = 12 minutes. No wonder lawyers have such fancy suits and a reputation for overcharging :)
378.
by Closi 3y ago
While invoices will get tracked to a cost-center in the accounting package, usually project managers won't access this directly and will have their own spreadsheets to the side to manage plan vs actual (i.e. they may also need to think
379.
by Closi 3y ago
Demonstrating you can do it yourself shows a level of investment and commitment to AI in your platform that integrating LLAMA does not. And from a corporate perspective, it means that you have in-house capability to work at the cutting-edge
380.
by Closi 3y ago
Thank god we aren’t talking about a model counting syllables then.
381.
by Closi 3y ago
I work in manufacturing/logistics and lots of facilities have starlink. With all software in the cloud now (e.g. WMS) losing connectivity can cause a facility to stop, so having an entirely segregated line on an entirely separate techn
382.
by Closi 3y ago
LLM's do that ok too... just not for crazy complex equations that would be tough for most humans with pen and paper. See below which I have just run on GPT4: https://chat.openai.com/share/3adb3aa2-8aec-474f-bdb0-4d
383.
by Closi 3y ago
This conversion is inherent in the problem of language and maths though - Two, too (misspelt), 2, duo, dos, $0.02, and one apple next to another apple, 0b10 and 二 can all represent the (fairly abstract) concept of two. The conversion to a
384.
by Closi 3y ago
I don't think that's really how it works - sure this is true at the first level in a neural network, but in deep neural networks after the first few layers the LLM shouldn't be 'thinking' in tokens anymore. The toke
385.
by Closi 3y ago
Yes, but that acknowledges that there is a difference between understanding how to perform reliable calculations, and actually being able to perform reliable calculations. Humans are good at the former, but not the latter.
386.
by Closi 3y ago
Humans were terrible at getting calculations right - that's why we invented abacuses, slide-rules, books of mathematical tables and tabulation machines.
387.
by Closi 3y ago
Of course if you only consider the disadvantages, something looks bad. The advantages of CSV are pretty massive though - if you support CSV you support import and export into a massive variety of business tools, and there is probably some f
388.
by Closi 3y ago
I’ve come to the opposite conclusion personally - AI model inference requires burst compute, which particularly suits cloud deployment (for these sort of applications). And while AIs may become more compute-efficient in some respects, the t
389.
by Closi 3y ago
I don't think that statement can be reliably applied if you are doing infosec at a large company.
390.
by Closi 3y ago
Current generation LLM's aren't about doing tasks that humans struggle with - they are about doing tasks that humans find easy quicker than humans can do themselves. Future generation AI's are highly likely to have more capab

More ›