Gemini's new discounted Flash pricing is still ~3x the price of GPT Luna, and that was released a week or two ago.

Everyone's hemorrhaging money here, but Google feeling they can't match or undercut isn't a great sign for them. And Gemini is flagging in intelligence too, which would be a strong reason to underprice if they could. Gemini Flash used to be your cheapest option. I'd say structural advantage still puts the odds in Google's favor in the long run, but has to be a worse situation than they ever expected to be in.

Investors are chasing a company that achieves some kind of AI singularity. Let's set aside the fact that there's no reason to believe there is anything but diminishing marginal returns by making marginal refinements to current frontier models. Let's imagine someone hits the jackpot and gets, not even AGI, but a system that's as far ahead of today's frontier model are ahead of 2020's GPT 2.0 in performance.

Globally AI revenues are 150 billion, against a cumulative burn rate of 450 billion. A model that is a generation ahead of others would almost certainly capture the lion's share of that revenue.

If AGI magically appears as a Sam Altman has promised investors it will, a hundred million is way too low. Add, maybe, another zero to the revenues.

Conservatively, a safer assumption is that frontier models will get marginally better based on refinements in training and reinforcement and the other bits and bobs that go into these systems. The nobody is winning the lion's share of anything, at least overnight. But you have to define "safe". By "safe" I mean unlikely to lose money. But some investors are clearly defining "safe" as "having the greatest chance of owning a piece of the biggest thing ever."

I'm not following this super-closely, but if Anthropic is pursuing adding multi-step model based reasoning to their system, that could be the basis of a generational leap in capability. But if that is an approach that looks like it has a chance of working, then their competitors are no doubt pursuing the same thing. In that case you'd expect the revenue pie to grow as the scope of model utilty increases, but that growth to be split among several competitors. This could credibly result in a revenue stream for some of them that is as big as the entire industry's revenue stream today. But there's going to be hell to pay on the data center impacts end of things.

Ideally, pay cold hard cash to opt in.

Second choice would be to provide some real add-on value to opt in.

What's really going to happen if opt-out becomes too toxic for twitch: They will go opt-in but enshittify the product for those who don't opt-in, touting a "better" (meaning maybe as good as August 2026) experience for those who opt in.

Sort of like the "opt in" toll roads in places where the free roads went from being decent to dangerous due to "lack of funding" to pay for maintenance.

They shouldn't be trying such a desperate solution without trying chlorox and ivermectin first.

Just don't leave the country after you participate.

For that matter, you may suffer remote retaliation even if you stay home. I would leave this to the military or other government branch.

Also there is the question of legality, government operation or not. It sounds to me like vigilante justice. And countries who like their anti-American cybercriminals will certainly treat it as illegal. (Hence the advice not to leave the country.)

We certainly need a way to stop cybercrime, international or otherwise. But this doesn't strike me as a good way to go about it.

Finally a prediction: in the not too distant future, one or more offshoots from this will turn into unmanageable domestic criminal organizations.

The parallels between this and letters of marque are scary.

It's not a remotely level playing field. One soldier who took part in the Venezuela raid basically had orders in hand when he placed multiple bets on the timing of the raid, netting a ton of money. There have been many other examples of inside information being used to stack the odds.

This is absolutely not true. There are plenty of people in the financial side of a business that have information that is material but non-public, and trading based on that knowledge is illegal, including sudden cuts in spending, audit results prior to publication, and more. HR will be aware of major layoffs. Facilities might be aware of impending site closures before they're announced. All these things can be used for insider trading.

Lower-level people aren't usually pursued because they're small enough to fly under the radar. But that doesn't mean it's not illegal, and it doesn't mean it never happens.

They're an advertising company, not a hardware company.

Don't forget the surveillance for profit revenue.

I'm reasonably confident there were other, previous known-at-the-time deaths from trebuchets but the records of their deaths or the reasons for the deaths have been lost.

https://www.youtube.com/supported_browsers?next_url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DCrbp8Y_yCIs

is not what you want

https://www.youtube.com/watch?v=Crbp8Y_yCIs

may be what you want.

The summary states that Reddit owes the OP a reason why the user was banned, as well as a complete, untruncated copy of the user's posting history.

The summary states that the relevant law is Article 17 of the EU's Digital Services Act.

You can bet that both black and white hats are poking around any popular tool that allows collaboration.

Teams, Discord, and others likely have high-impact bugs yet to be found.

If it has a high false-error rate the city may miss out on revenue and someone with a warrant may not be arrested, but any false-positive is an injustice to the person pulled over.

I'm willing to be the one-in-10,000 false-positive from a license-plate misread but not the one-in-100 (1% false positive) or 1-in-4 (25% false positive).

On the other hand I can tolerate my city missing out on ticket revenue or a missed warrant-arrest because only half of its license plates can be confirmed as "almost certainly correctly read."