Another Dream Views user? Rare to see nowadays. They went a bit crazy with the -ILD-isms, but it was a good time. I'm guessing they'd call this a VRILD.

I didn't see a usable demo oneironauts could try. Might be worth making some day.

Free version of Gemini also gets it right

This seems like a two wrongs don't make a right situation. Yes, the LLM providers are in the wrong for not vetting data. But in my mind, the person poisoning data is also partially responsible for negative downstream effects that are caused. What % of the blame is up for argument, but neither side is 0% in the wrong.

So the type and content of what fiction I write should be controlled so that LLM companies may choose for the sake of maximum profit to haphazardly collect any data they please and to neglect the safety and validity of the content they choose to serve users?

If it's meant to be fiction and not presented as a real-world fact, then I don't see any problem on your part. If you're presenting it as fact solely to trick an LLM into believing it, then yes I see a problem.

It is a common trope in some forms of fiction for the author to insist "this really happened I promise"

It is still meant to be understood that it obviously did not happen, despite such claims

We call it self defense when you fight back against a much better equipped attacker.

If you fire a gun in self defense and hit a bystander, you're still responsible.

In this case its more a, i stole a gun at a house burglared and sold it on the black market scenario. How about honest work and honest living?

If the ants take poisoned sugar tobthe queen, maybe they leave that house alone from now on?

Somewhat related, but I sicced an LLM at my wacom's BT and USB packets to connect it to an unsupported system and the dang thing got everything decoded in 15 minutes. Can't wait to resurrect some old hardware I've got laying around.

Incredibly pedantic, but I think that's describing a positive feedback loop where the previous result amplifies the next result further.

That was a year+ ago, which notably is when they decided to clean the game up. I don't see the point of using the past as an example when they very explicitly (like it was the very first set of criticisms aired in their keynote yesterday) are trying to move away from that.

What you're describing at the end is an attempt for a company to diversify and cutting their losses early to stem bleeding from projects which wouldn't have met the expected ROI. Gaming is a difficult industry and the majority of games do not get big. It's not surprising that the majority of their efforts to branch out failed.

None of your criticisms of the main game apply during the past year of updates. That + their attempt to create "sanitized" servers this year looks like another attempt to diversify - away from the microtransaction-based revenue source into a stable subscription source, assuming they get the playercounts.

That would likely not pan out well. I’m just talking about RS3 here, but their bot detection strictly works off behavioral characteristics - something that’s quite vulnerable to the current/upcoming bots which incorporate LLMs into the mix.

From what I could tell they have zero client-side detection like other, more competitive games. Maybe they wanted to pull out of that cat-and-mouse game? Their solution’s likely very domain-specific.

And bot detection isn’t a field where you can cleanly get a bot/human result. Especially with how bot-like some hardcore players get at the game. If it’s too strict you’ll get false bans, which is a support nightmare. It can’t be easy threading this line

Yeah, they are not good at it, they don't really have an incentive to filter out bots, they increase revenue by paying for the game and adding players. The players also like using modded clients like runelite, so they don't do their part either.

They have the ecosystem to test bot detecting/banning tech, but definitely not the tech. A pivot would require prioritizing bot detection over gaming, which is something that requires both management and players to mature beyond a game.

When I was playing, ~20 years ago, sometimes a real staff member popped up to talk with you when bot-like behavior was detected

It was pretty effective with people letting the bots run unattended (and annoying).

Of course that can't work as well now with LLMs.

I understand why they did it - medical emergencies and all. Have to pay the bill somehow. But yea, 0.47.05's still my go-to. Just can't vibe with the new UI/keybinds.

Old controls were on their roadmap for a long time, but the benefit from their perspective shrinks as time goes on. I doubt we'll see it at this point.

"It takes a ton of pressure to make a diamond". I read that in a game when I was a kid and took it as a guiding principle in my life.

Overcoming hard times absolutely made me a better person, but as I've grown I've seen many examples of the fallout you're referring to. Sometimes people just break.

You may like this old comment by dang who did this type of analysis: https://news.ycombinator.com/item?id=24215601 (2020)