Meta to build election operations center in Europe to inspect AI content
- Reference: 1709022610
- News link: https://www.theregister.co.uk/2024/02/27/meta_telection_infotrmation_ai/
- Source link:
As part of the social media giant's efforts to moderate content, Meta is working with the European Fact-Checking Standards Network – a project backed by an association of fact-checking organizations – to train people to inspect AI-generated content and other forms of digitally doctored media.
"We remove the most serious kinds of misinformation from Facebook, Instagram and Threads, such as content that could contribute to imminent violence or physical harm, or that is intended to suppress voting," it [1]explained on Monday.
[2]
"For content that doesn't violate these particular policies, we work with independent fact-checking organizations – 26 partners across the EU covering 22 languages – who review and rate content. We are currently expanding the program in Europe with three new partners in Bulgaria, France, and Slovakia."
[3]Election security threats in 2024 range from AI to … anthrax?
[4]OpenAI latest to add 'Made by AI' metadata to model work
[5]AI political disinformation is a huge problem – but harder to fight than ever
Meta already has systems in place that label realistic AI-generated images created by various tools from Google, OpenAI, Microsoft, Adobe, Midjourney, and Shutterstock uploaded on its social media platforms. Nick Clegg, the group's president of global affairs, [6]previously announced that the label "Imagined with AI" will be added in the next few months.
Advertisers have to be transparent if their campaigns on social issues, elections, or politics, contain synthetic content under [7]new rules . Other types of users will soon be able to disclose when they share AI-generated images, video, or audio in upcoming changes. Meta warned that it may apply penalties if people aren't honest.
[8]
Meanwhile, the content moderators will inspect media and can tag it with further information – like attaching an "altered" label, which will cause such medias to be treated as less relevant, so it appears lower down on people's feeds. They can also add more context and information to avoid confusing users. Misleading adverts that get debunked by fact-checkers will be removed.
Adverts aimed at suppressing voter turnout, undermining the election process, or confusing the results are not allowed on Meta's platforms. Unfortunately, 95 percent of people don't bother reading the extra information on fact-checked posts, according to Meta, so more stringent action was necessary.
[9]
Advertisers are also now required to disclose the source of their funding – netizens will see a "paid for by" label.
The latest changes come as Meta and other AI developers face increasing pressure from regulators to ramp up content moderation practices. Generative AI is rapidly improving, and commercial tools are available to anyone to create realistic synthetic media. The risks of spreading political misinformation are particularly high as the EU, UK, and US conducts elections this year. ®
Get our [10]Tech Resources
[1] https://about.fb.com/news/2024/02/how-meta-is-preparing-for-the-eus-2024-parliament-elections/
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2Zd3A10-sXZ8HhC9tuKDrtQAAAY8&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[3] https://www.theregister.com/2024/02/18/election_security_threats_2024/
[4] https://www.theregister.com/2024/02/08/openai_is_adding_a_watermark/
[5] https://www.theregister.com/2024/01/17/ai_political_disinformation/
[6] https://www.theregister.com/2024/02/06/meta_ai_label/
[7] https://www.theregister.com/2023/11/29/meta_ai_ad_disclosure_requirement/
[8] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44Zd3A10-sXZ8HhC9tuKDrtQAAAY8&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[9] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44Zd3A10-sXZ8HhC9tuKDrtQAAAY8&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[10] https://whitepapers.theregister.com/
You can't beat it
I can easily imagine that the very large majority of all comments, articles etc will be written by entire botnets controlling AI chat bots..
From the governments to the mafia, everyone will be trying to manipulate the poor souls who rely on social media for information.
"Fact-checkers will label AI-generated media for upcoming EU elections"
What could possibly go wrong? After all, it may be artificial, but it's intelligent ! *
* possibly, more intelligent that those who fell for this?
.
Re: "Fact-checkers will label AI-generated media for upcoming EU elections"
What could possibly go wrong?
Think also on what can go right. Like the prospect of throwing Zuck in jail for spreading misinformation if/when his 'fact checkers' are shown to be wrong. But it seems to be reliant on this lot-
https://efcsn.com/governance-body/
with a UK rep from this lot-
https://fullfact.org/about/funding/
Ultimately, it is for our audience to judge for themselves whether we succeed
Except the whole point of most of these neo-fascist censor bots is to make sure audiences don't get to judge for themselves because they're all pushing for auto-take downs of anything they disagree with, along with punishment. The amusing thing about fullfact is they're developing AI 'tools' to automate censorship, so AI vs AI. And thus the AI wars begin!
Re: "Fact-checkers will label AI-generated media for upcoming EU elections"
I'm really looking forward to the increase in the chocolate ration!
Will this be like the time they fact checked the Hunter Biden laptop story and deemed it to be Russian information despite it turning out to be entirely 100% factual?
Expect this to benefit anybody pushing to enlarge the state and nobody else.
Expect this to benefit anybody pushing to enlarge the state and nobody else.
But.. but.. it's sustainable. It's also a bit of a circular firing squad with public money going to private 'fact checking' outfits, like "Fullfact". It's already a big business, and growing as governments impose more censorship legislation. I can't help thinking the money might be better spent on the education system to teach kids critical thinking and developing their own BS detectors.
There's a neat example on Fullfact 'fact checking' energy price increases, especially linked to British Gas profits. It correctly seperates BG and Centrica, but kinda glosses over how energy companies use financial engineering to move profits around. BG made a modest profit, Centrica made a much, much larger one, all thanks to our misguided energy 'policy'. It also glosses over the big increase in standing charges, and why continuing policies like this is only going to make inflation and cost of living problems worse.
No, inflation is our fault for spending too much. And the recession is also our fault for not spending enough.
Take Care ...... AIMind Minefields Ahead
Who now dares lose all credibility and say the machines are not winning?
And you surely know that all makes perfect common sense.