Gio Van Hoven

Be careful what you wish for. While I'm sure it feels good and validating to have this called copyright infringement, and be compensated, it's a mixed blessing at best. Remember, this also means that your works will owe compensation to anyone you "trained" off of. Once we accept that simply "learning from previous copyrighted works to make new ones" is "infringement", then the onus is on you to establish a clean creation chain, because you'll be vulnerable to the exact same argument, and you will owe compensation to anyone whose work you looked at in learning your craft. This point was made earlier in this blog post: https://blog.giovanh.com/blog/2025/04/03/why-training-ai-cant-be-ip-th… HN discussion of the post: https://news.ycombinator.com/item?id=43663941

Awful headline you added. https://blog.giovanh.com/blog/2023/03/17/replika-your-money-or-your-wi… explains: But this is a product that’s targeted at and specifically addresses the needs of people who are lonely and thus specifically emotionally vulnerable, which should make it worse to inflict suffering on them and endanger their mental health, not somehow funny. ... As cringe worthy as “AI girlfriend” is, there’s no question that’s what they sold people. They advertised it, they sold it, and they assured users it wasn’t going away even while they actively planned on killing it.

There's a pretty good retrospective written up on this blog[0]. In short: originally the purpose was nothing more "this account belongs to the person they claim to be and we've directly verified this with them". Unfortunately, people habitually misinterpreted the checkmark as not just being verification but also a tacit endorsement of the account by Twitter the company. Which isn't great when you get a high profile controversial event and it's lead organizer has a verified Twitter account. After that, they appended an "in good standing" qualifier, and it quickly devolved into a "you know a person who knows a person who knows a person" situation since they also announced a public pause of the program. (Notably, the ID check, while it existed, was pretty much abandoned. Twitter at some point began demanding ID scans to report things to their support, but that obviously never actually translated to a blue check.) Musk's version of it is hilariously simplistic, but also robs it of any and all value: just pay money for it and you'll get it. It works in the sense that it confirms the poster has a bank account (although this probably doesn't confirm much in and of itself), but any and all value of said verification is minimal because any old hack/scammer can do that. Verification is a difficult system to get right and people have all sorts of pre-baked in ideas on how it should work versus how it actually works and the use of a checkmark played a part into how Twitters version…

Homestuck is a phenomenon that I see at the periphery of my experience on the internet… it’s always a source of interesting reading material about internet culture even for those of us who never participate. This followup article has some interesting speculation about a key reason for the Kickstarter’s failure: https://blog.giovanh.com/blog/2021/01/14/more-on-the-hiveswap-odd-gent… > … Andrew and The Odd Gentlemen committed to a development schedule that started with a design document Andrew needed to write. However, he failed to produce that document on time, which wrecked the schedule. > … This period of limbo lasted for nearly two years of development time because Andrew wouldn’t deliver the cornerstone design document. … > Eventually, TOG decided they had wasted too much time making comic assets for Homestuck and needed to work on [King’s Quest] To me it looks like Andrew Hussie successfully created a moderately successful IP, but never developed the necessary management skills to run a business based on that IP of that scale.

Because training costs are sky-high, and handling an individual request still uses a decent amount of energy even if it isn't as horrifying as training. Plus the amount of requests, and content in them, is going up with stuff like vibe coding. If you want to know more about energy consumption, see this 2 part series that goes into tons of nitty-gritty details: https://blog.giovanh.com/blog/2024/08/18/is-ai-eating-all-the-energy-p…

Would you accept weaponized trademark infringement as a sibling to patent law? The behavior described here seems odd -- I wouldn't call it "good" or "bad" in a moral sense just technically impressive: https://blog.giovanh.com/blog/2023/10/26/apples-trademark-exploit/ and https://www.theverge.com/2020/9/14/21436760/us-customs-statement-onepl…

Sometimes, doing something is worse than doing nothing at all. > For that matter, using licensing to restrict government abuse obviously isn’t going to work, because licensing relies on state enforcement of rights anyway. If the state itself is compromised, having a clever little license isn’t the failsafe the OES thinks it is. It’s really just asking the police to arrest ICE with extra steps. This isn’t hypothetical, this is already codified in the law: even the international treaties that govern copyright have explicit exemptions for government action related to security interests. A license like this one can only ever possibly do harm. https://blog.giovanh.com/blog/2021/10/29/ethical-source-is-a-crock-of-… And if you want something more old-school: > It is worse than ineffective; it is wrong too, because software developers should not exercise such power over what users do. Imagine selling pens with conditions about what you can write with them; that would be noisome, and we should not stand for it. Likewise for general software. If you make something that is generally useful, like a pen, people will use it to write all sorts of things, even horrible things such as orders to torture a dissident; but you must not have the power to control people's activities through their pens. It is the same for a text editor, compiler or kernel. https://www.gnu.org/philosophy/programs-must-not-limit-freedom-to-run.…

Imagine if Apple sold OEM parts at a reasonable price. Imagine if they weren't pairing OEM parts to prevent repair shops from parting out broken devices. Imagine if they weren't making deals with manufacturers to prevent the sale of standard parts to third parties - even parts with no Apple IP like the ISL9240 [0]. Imagine if they weren't abusing Customs and Border Patrol powers to prevent distribution of legitimate parts via false acusations of counterfeiting [1]. Maybe then you could get a good screen from a small repair shop. It's not about margin. It's about parts availability. Apple stores are few and far between, but Apple products are ubiquitous. Official Apple repairs have them ship the device across the country to repair shops, which takes a lot of time. And they don't do things like component-level repair (i.e. specific chips, resistors, and other board components). They just replace the entire board and charge you for the new one. An independent repair shop doesn't have to compete on price. They can compete with fast turnaround, minimal travel, and component-level repairs. Or at least they could if they could get the parts. [0] https://www.youtube.com/watch?v=lTpHa70DDX0&t=200 [1] https://blog.giovanh.com/blog/2023/10/26/apples-trademark-exploit/

> [citation needed] Sure, here ya go: https://andymasley.substack.com/p/individual-ai-use-is-not-bad-for https://blog.giovanh.com/blog/2024/08/18/is-ai-eating-all-the-energy-p… https://blog.giovanh.com/blog/2024/09/09/is-ai-eating-all-the-energy-p… The first comprehensive environmental audit and analysis performed in conjunction with French environmental agencies and audit environmental audit consultants, which includes every stage of the supply chain, including usually hidden upstream costs: https://mistral.ai/news/our-contribution-to-a-global-environmental-sta… https://andymasley.substack.com/p/for-the-climate-little-things-dont > Disingenuous strawman. Tech CEO's and the like have been exuberant at the idea that "AI" will replace human labor. The entire end-goal of companies like OpenAI is to create a "super-intelligence" that will then generate a return. By definition the AI would be performing labor (services) for capital, outcompeting humans to do so Isn't that literally the selling point of software, performing something that would otherwise have to be done by humans, namely both keeping calculating research, locating things, transferring information, and so on, using capital instead of labor, transforming labor into capital, and providing more profits as a result? > Unless OpenAI wants it to just hack every bank account on Earth and transfer it all to them instead? Or something equally farcical It's extremely funny that you…