dehrmann

For this particular case, it's probably worth it because a dermatologist visit is $200. If I were the NHS or Kaiser, I'd partner with other integrated insurance and service providers to collect as much training data as possible and fund classifier development.

This is very wabi-sabi.

Not SaaS, but I had an idea for using a CPU emulator to run a compiled binary under Java, with libc implemented in Java so standard file IO works. I punted on the idea because it seemed pretty difficult and maybe take six months. I recently gave Claude a Java RISC V emulator and a partial bridge between Java and libc, and in under an hour, it could run bash and sqlite from Debian.

I have no idea how correct the code is or what I'd do with this, and I have no plans to publish it for those reasons.

For all the hand wringing we're doing over AI, I'm surprised that post-covid, regardless of origin, there wasn't a more serious discussion on what sorts of biological research are worth the risk.

It's possible to get billed for things not exposed in the UI. ~10 years ago, I had one of these. A Glacier upload chunk was stuck in a staging area for over a month. I couldn't complete or cancel the upload, or remove the chunk. Support was able to, and they refunded the money without much hassle. But I did have to reach out to support (as a <$100 per month customer) to ask. Glacier was new, so I chalk it up to it being a new product, not anything greedy or malicious.

You can get away with a lack of rigor in software engineering (from both humans and AI agaents) because it's often cheap to try it and see. Once you get to operating the platform, it changes, and operational decisions need to be made carefully and code needs to be deployed carefully.

The issues are they likely have seeds, might not travel or store as well, and are smaller than consumers are used to. There's also already a supply of table grapes you'd be competing with. To your experience, if they're ripe, they'll taste sweet and grapey. They're not that different.

I'm being very mindful of Gell-Mann Amnesia and chatbots. They speak authoritatively and are right often enough, but I've had enough cases of them saying very incorrect things in areas I know well that I have to remind myself that those cases aren't unique.

This is why you read the code, not the documentation.

Not sure if it was intentional, but saying "uses AI" in the headline makes it sound like this is a modern thing, but the techniques here are closer to the linear regression side of AI than the trasnformer side. The headline could have been "DraftKings Targets Problem Gamblers." It would have put the outrage where it should be.

Most places at this scale have code freezes in place well before conferences. The most likely issues are some launch couldn't handle the scale or periodic deployments have been saving them from some sort of long-standing leak bug, and pausing going into Dreamforce meant some service hasn't been restarted in a week. Historically, Salesforce sharded by customer, so that goes against both of these, unless it's in a routing layer.

I think you're thinking of something else, but they also added "capabilities" to Hack (their main frontend language; it's a PHP fork) to make leaking data between requests or in logs painful.

When I interviewed at Facebook in 2019, one interviewer was on an integrity team. I'm not sure if it was his take or what the org settled on, but it's that Cambridge Analytica wasn't Facebook's fault (users willingly granted access), but it was their problem. Not surprising some of the same techniques got used by others.

> you will face criminal and financial penalties for damages caused

Even with where we're at right now, if people involved with the HF or Ruby Gems attacks faced legal consequences, the industry would get a lot more conservative very quickly. And unlike social media, section 230 doesn't apply, and there aren't any free speech concerns.

> load-bearing

I'm worried I'm going to start picking up claudisms, and then accused of being AI.

> HF should sue them

HF, like the Nvidia subsidiary?

> where the performance carries a lot of the meaning

The big one for me is you don't necessarily know Elizabethan slang, but when you see the play performed, the dirty lines get delivered dirty.

We're going to see more of this as the web 2.0 and mobile development booms fade, and I don't have a good solution. What do you do when you're a mature SaaS with a solid customer base, stalled growth, headcount for hypergrowth, and investors looking to move on.

Even shrinkwrapped software is subscription-based. It's just whether the subscription is monthly or every 5 years.

Clicking through is worth it just for the "draw an svg of a circle" bit.

> since 2019

At that point, why wouldn't you pick a major cloud vendor? This feels like more of a governance failure.

The buyers are saying it's because the human-created models are better. This might be true, but it could also be some other things. Maybe they're doing the AI dream and using AI to create the models themselves. The article cites revenue maybe AI model are selling, just cheaper. Maybe human creators are buying from humans out of solidarity.

These are the current ingredients:

> INGREDIENTS: SUGAR, GLUCOSE SYRUP, COCOA BUTTER, FULL CREAM MILK POWDER, COCOA MASS, SUNFLOWER OIL, SKIMMED MILK POWDER, LACTOSE, WHEY POWDER (FROM MILK), FAT REDUCED COCOA, MILK FAT, BARLEY MALT EXTRACT, SALT, EMULSIFIER (SOYA LECITHIN), EGG WHITE POWDER, PALM FAT,

There's nothing particularly weird in it. Soy lecithin is probably the weird one, but it's in a lot of plain chocolate bars, and is pretty benign.

It's not. 23,000 jobs is 0.01% of the workforce, so it's negligible in terms of lost sales (probably hurts consumer confidence more). But it is a signal that the economy isn't running too hot, so the company's much larger borrowing costs aren't going up.

It's more like Buck and Tupperware.

How do models approach SVG generation? In one version, I imagine them actually trying to reason about them as an LLM. In another, I imagine something closer to a GAN.

A nit, but fusefs is a well-known thing, so you'll face some naming confusion.

I enrolled in 2002. The Nasdaq bottomed in September 2002, there was talk everywhere about outsourcing, but I went into it because that's where my passion was. I graduated into the 2006 job market which wasn't booming, but recovered enough. I remember chatting with younger coworkers about how we wish we saw the dot-com bubble. Then I saw the 2010-now tech industry.

Turns out I bought what's probably the largest investment in my lifetime at the bottom.

The industry will change in the next four years. What I didn't realize in 2002 was widespread broadband internet would arrive and change everything. Revision to mean is a thing, and I wouldn't expect the 2010's boom where tech moved to take over every corner of our lives, but it will still be there, and the systems take proper care and feeding. Remote work is equalizing global tech wages, but the talent's gotten better and the pie's gotten bigger.

If you don't have a passion for tech, pick a different major because you'll spend the next few years questioning your choice. If you stay with it, look for other areas in your life you enjoy so you can plan out a minor or double major. If you plan it right, you'll be able to reassess in three years and possibly pivot to math, actuarial science, electrical engineering, or mechanical engineering with only an extra year.

It's like a basic Kindle. Or Ryanair.

Between this, the proprietary nut patent, and the heated seat subscription, I'm probably not going to look at BMW for my next car. They've been going back on things, but all of this adds up to they don't get it.

It's differently interesting. It's interesting that they didn't get an important detail right with a partner, so more on Anthropic's attention to important details rather than the power of their models.