waldrews

13 ms·

121.
by waldrews 2y ago
Is there a list (blog, Twitter feed) of such stories of AI embarrassments? Might make a good teaching resource for explaining to a less technical audience why many of us still have to work hard to make AI work reliably for serious applicat
122.
by waldrews 2y ago
How does it compare with LinqPad, polyglot notebooks, csi.exe?
123.
by waldrews 2y ago
The root problem is that there is no good way to bill a modest surcharge on top of OpenAI etc. usage costs. GPT store monetization (if they ever get around to it) would be a partial solution. But anything that depends heavily on LLM'
124.
by waldrews 2y ago
Ok, I hear a lot of 'AI is a superset of ML' definitions, and yes, that's historically true, but today AI is shorthand for LLM's and others (image generation, embedding, DL-based time series). I'd draw the line at
125.
by waldrews 2y ago
Some of us looked at it and decided the results were so bad it was a net loss to even try, but it was just so cool...
126.
by waldrews 2y ago
Oh great. Like we didn't have enough 'this changes all our architectural tradeoffs and business models' events last week.
127.
by waldrews 2y ago
Cool how we're far enough into the adoption cycle that 'I reduced use of LLM' is a mainstream thing to hear among sophisticated techies.
128.
by waldrews 2y ago
Would this be suitable for a VPN-like setup with some remote servers and distributed home dev machines? In other words, does the 'homelab' in the title imply LAN-specific functionality?
129.
by waldrews 2y ago
The world needs better pricing models for such tools. Monthly fee, even a cheap one, is just too much of a congitive load. Maybe a subscription service for many SaaS that allocates fees to its components according to a usage based formula
130.
by waldrews 2y ago
The pricing is still such that you can't routinely use a customized Gemini with your own long fixed pre-filled context + a variable short query. If the one-time compute cost of the caching could be effectively amortized over many quer
131.
by waldrews 2y ago
Aargh, painfully verbose GPT-generated filler nonsense. Naively I thought we were safe from it on HN.
132.
by waldrews 2y ago
Exactly. In the long run we want a cheap/fast LLM layer that we can call millions of times in a search/conversation-with-itself pattern. It may be that we'll interact with it the same way, and GPT-6 will be lots of GPT-5&#x
133.
by waldrews 2y ago
What's static/dynamic compute in this context?
134.
by waldrews 2y ago
I think he's referring to a 'zero interest rate phenomenon,' a shorthand for an excess that happens when capital is cheap and complexity is rewarded and too easily funded. Supposedly, this leads engineers to choose overly co
135.
by waldrews 2y ago
Oh yeah, definitely reasonable. I'm a statistician by trade, so if I see R^n, I want to put a multivariate normal measure on it, which means hyperspheres. Hypercubes are cool too. Just as long as we don't treat the limit from g
136.
by waldrews 2y ago
You're using a shorthand that captures an important point but does not quite dispose of the objection. Let R(X) be an indicator function that's 1 when largest root is real and 0 otherwise; then we want E[R(X)]. I think your poin
137.
by waldrews 2y ago
Meaning Space transcending the imperfect Language Space? Yes, there's been some recent thinking in this direction, e.g. Zhuangzi, "Words exist because of meaning. Once you've gotten the meaning, you can forget the words. Whe
138.
by waldrews 2y ago
Serve them on a firewalled port, then: 1) VPN if you need to expose them to multiple trusted users, 2) firewall rules to make them accessible to your IP range, or (probably easiest), 3) access them by ssh tunnel.
139.
by waldrews 2y ago
What's wrong with VSCode in this context?
140.
by waldrews 2y ago
Does anyone has recent experience with Maui targeting mobile? It had a not-quite-caught-up-to-Xamarin reputation for a while, has that changed now that we're on .Net 8?
141.
by waldrews 2y ago
Could this be used in Windows by exposing a Samba file share from WSL or Docker?
142.
by waldrews 2y ago
Can't really speak to Java, but ASP.Net, in its new open-source friendly incarnation, is perfectly viable for modern development (and there's still nothing quite like Blazor Server). They have a good story for integrating the new
143.
by waldrews 2y ago
I think the lesson here is obvious. Who among us hasn't been terrorized by PhysEd teachers, forced to run laps or climb thing or something against our will? Athletic directors must be even worse. Ban athletic directors!
144.
by waldrews 2y ago
Statistics Ph.D. here. This phenomenon is the scourge of our existence in the big-data age. It's why we have endless stories about miracle medical diagnostic methods that can tell everything about you from a retina image or a recordi
145.
by waldrews 2y ago
The world needs a way to pay for thin AI wrappers without having to turn them into subscription services. There are lots of potential applications where an indie developer can add value on top of LLM's, but can't subsidize GPT-4
146.
by waldrews 2y ago
Why don't we see more general purpose computing environments where the whole interpreter state is persistent and serializable by default? The only ones I'm aware of is R's .RData mechanism, some Smalltalk environments, and s
147.
by waldrews 2y ago
Yeah, that's kind of weird - Apache, except totally not Apache. They probably meant to do a source-available license with free non-commercial use allowing inbound contributions - makes sense for a startup, but... please get a lawyer a
148.
by waldrews 2y ago
To clarify: is the referenced advice (public wifi bad, change passwords) pedantic and/or wrong, or is the statement (this advice is outdated) pedantic and/or wrong?
149.
by waldrews 2y ago
The problem here isn't that an LLM hallucinates. The problem is that nobody asked for an AI response, and Meta pushed content to a forum that makes such claims, which could easily mislead or at least confuse people not sophisticated e
150.
by waldrews 2y ago
Suggestion: have GPT-3.5 also write out unit tests, both strings that should pass and ones that should fail, and run the JS version of the regex on that in-browser.

More ›