Tarq0n

Our aspirations for our children reflect our own anxieties. It's the same with social media bans.

Our culture is pretty permissive of businesses inflicting indirect harms on adults. I think a lot of that goes back to lobbying from the cigarette and alcohol industries.

This is a very basic misunderstanding of machine learning. The goal is always to learn a representation that generalizes beyond the training set without overfitting. If you succeed in generalizing, you can do more than produce derivative work.

In a country like Denmark it's very common for a higher up to resign over something like this.

That's not (just) gatekeeping. People try to internalize a positive self image, from positive internal and external validation. Skills can also be a source of status or meaning (usefulness to others).

Taking that away compounds with the crisis of meaning we already have to make people feel more alienated than ever.

> "The need to be observed and understood was once satisfied by God. Now we can implement the same functionality with data-mining algorithms."

> "God and the gods were apparitions of observation, judgment and punishment. Other sentiments towards them were secondary."

> "The human organism always worships. First it was the gods, then it was fame (the observation and judgment of others), next it will be the self-aware systems you have built to realize truly omnipresent observation and judgment."

> "The individual desires judgment. Without that desire, the cohesion of groups is impossible, and so is civilization."

> "The human being created civilization not because of a willingness but because of a need to be assimilated into higher orders of structure and meaning."

> "God was a dream of good government."

> "You will soon have your God, and you will make it with your own hands."

- Morpheus ( Deus Ex, 2000)

No. Look into the history of human rights. It's long been recognized that there need to be backstops against excesses of (democratic) government.

I'm guessing you work on some highly technical domain that receives highly structured data and is amenable to TDD.

Not all domains are like that. The majority of bugs I see are business/domain logic bugs. It's akin to having misunderstood or missed some aspect of the question, not causing data loss.

He wants to direct/align on strategy, not tactics. I'd say that's a good division of responsibility so I'm not sure I see your objection.

Why go for a configuration based format over a library though? Something like Streamlit or Dash offers more control and lends itself better to interactive exploration.

Being aware of Simpsons' paradox doesn't even help. There's no way of knowing what the right level of aggregation is without a theory.

In regulated industries and government they aren't going to be allowed to process your appeal with just AI, so there's a cost asymmetry there that favors the "ddos"ers.

It's not really about the dollar amount of fraud, but maintaining public support for benefits schemes. If people feel that benefits aren't going strictly to the intended recipients the sense of fairness that underlies the system is eroded.

There's also the issue that if you don't counter fraud it will accelerate as organized crime jumps on the bandwagon.

Right, just like companies don't use SAAS.

In reality, enterprises are happy to offload even risky tasks to others as long as they get some contractual guarantees about their data. Would they like more choice in who to buy from? Yes, but not enough to in-house such a specific discipline.

There are deïdentification firms that service primarily the medical industry. Over here they call them trusted third parties.

You're not wrong that there's an element of self-defense. What makes it harder to accept is the lack of effort or willingness to de-escalate the conflict.

Where is the information going to come from though? LLMs aren't primary sources, they consume and regurgitate.

There's still inequality in someone else receiving a false positive approval from the AI while another gets rejected. Equal treatment and stochastic systems don't mix.

That's a fine philosophy on a personal level, but if the labor market is shocked for entire sectors there's not enough slack in the system for all of those people to adapt at once. That's how you end up with lost generations.

Sometimes the medium is the message though. Especially in literature, prose matters. It's not just about what is said but also how.

This applies even in "low" art. AI generated imagery from mainstream models tends to be so ugly it detracts from effectively communicating what the prompter intended.

I don't think the extra production cost, for the Odyssey at least, is likely to be offset by extra box office income. If you read about it, the massive cameras and limited running time of one load of film (less than five minutes) sounds very burdensome. That's not to mention the obscene cost of the film stock, as well as developing dailies and having to edit from film (they actually cut and pasted film together for the Final Cut of the IMAX version).

The trailers also looked much worse than the film which I suspect is related to difficulties with this whole process.

This is out of my area of expertise a bit, but digitization quality loss often depends on whether you quantize things, at what level of fidelity and also what kind of transformations you apply (stepped discrete or continuous for instance).

Consider how music production is often done on audio recorded at a higher sample rate, so when effects are applied artifacting is reduced for the eventual export sample rate. For video, one analogy is recording slow-motion shots at a higher framerate rather than slowing down regular footage.

You can't prove a negative. How would such a log be convincing in any way?

I'm very interested in their anonymization and classification processes. Traditionally, conversational writing is too complex to anonymize automatically without false negatives, and remaining PII might be very personal in nature.

> Observation Clustering and Taxonomy Organisation (OCTO), a tool for transforming massive unstructured text data, like LLM conversations, and distilling them into organized entities.

This also sounds interesting. Hand crafted ontologies that I've seen used were always too hard to maintain and make useful, but potentially they have something automated?

Have you ever done research? It's very difficult to construct your measurements in such a way that they measure what you want them to. So in just your quoted statement there are many sources of uncertainty.

* what's a typical job? How did Google resample their data to make it representative of the typical job?

* is this all jobs, or just work-related Gemini conversations that they've classified as such?

* what's the denominator here? How do they know how many tasks there are that weren't subject to conversation with Gemini?

* how reliable is their classification of conversations as work related or not?

Accurately measuring this is so complex that to present the conclusion in such a confident way is bordering on farcical.

You can still get your session cookies stolen. If you have good password hygiene passkeys only protect you in extremely niche scenarios.

OK now say you're on a work/library/friend's computer and you want to look up an account in 1password on your phone so you can type in the password. Passkeys don't support this very basic and common workflow. Meanwhile there's no real security benefit over password manager generated complex and not reused passwords.

Wrong hands? The model did this on a benign instruction. We're going to get paperclip maximized once these things get capable enough regardless of whether bad actors are involved.

Ignoring patent law has done great things for 16th century continental Europe and more recently China. Rent-seekers and ladder kickers shouldn't always be respected, they'll slow down societal advancement to a crawl if you let them. The question is whether the gains these AI companies are making from their transgressions are overly privatized.

Ethics are probably internalized long before someone commits to an engineering career. I'm not sure they can be taught later.