Meta plans to replace humans with AI to assess privacy and societal risks

[highplainsdem]

Source: NPR

For years, when Meta launched new features for Instagram, WhatsApp and Facebook, teams of reviewers evaluated possible risks: Could it violate users' privacy? Could it cause harm to minors? Could it worsen the spread of misleading or toxic content?

Until recently, what are known inside Meta as privacy and integrity reviews were conducted almost entirely by human evaluators.

But now, according to internal company documents obtained by NPR, up to 90% of all risk assessments will soon be automated.

In practice, this means things like critical updates to Meta's algorithms, new safety features and changes to how content is allowed to be shared across the company's platforms will be mostly approved by a system powered by artificial intelligence — no longer subject to scrutiny by staffers tasked with debating how a platform change could have unforeseen repercussions or be misused.

-snip-

Read more: https://www.npr.org/2025/05/31/nx-s1-5407870/meta-ai-facebook-instagram-risks

[Walleye]

1. The modern version of the great replacement theory

[highplainsdem]

3. Maybe Zuckerberg is wondering if he can eventually achieve what OpenAI CEO Sam.Altman says is

the new Silicon Valley dream - a company worth a billion dollars or more with just one human (founder/owner/CEO) and everything done by AI. Get rid of all those pesky other people who not only want things from the company (salaries, health insurance, fair treatment) but might be whistleblowers and/or write tell-all books.

[SheltieLover]

6. Repukes dream

[highplainsdem]

9. True.

[Walleye]

14. Some owners of big companies think of payroll as highway robbery

[bucolic_frolic]

2. Does posting ambiguities fvck them up?

AI is constantly telling me things that aren't true. It tends to have logical fallacies ... universal generalization, which may or may not be applicable in your situ/local/problem.

[highplainsdem]

4. Those are hallucinations, common with LLMs or generative AI, and almost anything can fuck

them up.

Which is why, for example, it's disastrous that the Trump/Musk regime wants much of the govermment automated to run on Musk's AI.

[bucolic_frolic]

10. Yes, hallucinations are what they've labeled them in an effort make them human

I don't buy the term but have probably lost the war.

[ananda]

11. This will have disastrous results on every possible level.

But that's OK, I presume, because it helps us avoid
giving jobs to "them."

No DEI whatsoever.

[Bengus81]

5. Pffttt...who cares? This will be like the tree falling in the forest and not making a sound

I won't be around your FarceCrook site so do what you want to do Fuckerberg.

[highplainsdem]

8. True. Facebook is already.drowning in AI slop and bot posts.

[Martin68]

7. What could go wrong?

[Progressive dog]

12. AI is certainly artificial but I doubt

that it will ever be intelligent. The machines can't fact check, they can't even observe. Every thought they have is based on what information in the databases and algorithms that define how the information is used.

[PatSeg]

13. I didn't think Facebook had any real live humans anymore

It all seems to be run on algorithms and contacting an actual person is nearly impossible. As a company they've just gotten worse and worse in recent years.

[William Seger]

15. That's exactly the kind of thing AI doesn't do well

AI bots are so good at imitating how humans talk that it's easy to forget that they have NO ability to actually use logic correctly, because: (1) they are incapable of independently fact-checking any premises to their logic; and (2) they are incapable of recognizing whether an inference is logically valid, in the sense that if the premises are true, the conclusion cannot be false.

[Wonder Why]

16. In reality, replace employees so the execs can get bigger bonuses.

[ProudMNDemocrat]

17. The same can be said of Phone Maze menus in order to speak to a human..

One calls a number and gets a menu instead.

For elder Seniors with cognitive and hearing issues, a nightmare. I have neither at the moment, but I find navigating through menus frustrating as well. Humans cost too much.