


Google Used AI To Suspend Over 39 Million Ad Accounts Suspected of Fraud (techcrunch.com) 25
An anonymous reader quotes a report from TechCrunch: Google on Wednesday said it suspended 39.2 million advertiser accounts on its platform in 2024 -- more than triple the number from the previous year -- in its latest crackdown on ad fraud. By leveraging large language models (LLMs) and using signals such as business impersonation and illegitimate payment details, the search giant said it could suspend a "vast majority" of ad accounts before they ever served an ad.
Last year, Google launched over 50 LLM enhancements to improve its safety enforcement mechanisms across all its platforms. "While these AI models are very, very important to us and have delivered a series of impressive improvements, we still have humans involved throughout the process," said Alex Rodriguez, a general manager for Ads Safety at Google, in a virtual media roundtable. The executive told reporters that a team of over 100 experts assembled across Google, including members from the Ads Safety team, the Trust and Safety division, and researchers from DeepMind. "In total, Google said it blocked 5.1 billion ads last year and removed 1.3 billion pages," adds TechCrunch. "In comparison, it blocked over 5.5 billion ads and took action against 2.1 billion publisher pages in 2023. The company also restricted 9.1 billion ads last year, it said."
Last year, Google launched over 50 LLM enhancements to improve its safety enforcement mechanisms across all its platforms. "While these AI models are very, very important to us and have delivered a series of impressive improvements, we still have humans involved throughout the process," said Alex Rodriguez, a general manager for Ads Safety at Google, in a virtual media roundtable. The executive told reporters that a team of over 100 experts assembled across Google, including members from the Ads Safety team, the Trust and Safety division, and researchers from DeepMind. "In total, Google said it blocked 5.1 billion ads last year and removed 1.3 billion pages," adds TechCrunch. "In comparison, it blocked over 5.5 billion ads and took action against 2.1 billion publisher pages in 2023. The company also restricted 9.1 billion ads last year, it said."
That's great, but... (Score:2)
Please release the names of all the companies committing fraud or having fraud committed on their behalf. Lets name and shame them, show them this type of behavior isn't going to be socially allowed.
It's pitchfork time!
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
Good luck. They have names like BRJMKL and MAQEVN. These same shysters create fake-named companies on Amazon all the time.
Click disparity (Score:2)
It may be that the non social media sites have lost enough traffic now that identifying a high fake click rate site is much easier for Google.
Re: (Score:2)
While click fraud is an issue, it's only one of many.
the top five policy violations related to financial services, trademark misuse, ad network abuse, personalized ads, and gambling and games.
Scammers have won (Score:2)
Try to download an ISO of clonezilla
Re: (Score:2)
I imagine the typical user would search "clonezilla ISO" in google, so I did that, the first 2 links take me to clonezilla.org/download and the third link to sourceforge.net. All links seemed accurate. The download speed is slow with the mirror it gave me but it works. (I did not check the checksums.) What was your experience?
Re: (Score:2)
I just highlighted their text, chose search, and Google had the right site at the top of the list.
Would make sense. (Score:4, Interesting)
Re: Would make sense. (Score:2)
Accuracy? (Score:1)
And we'll never know the false positive / false negative rates.
So what google is saying is... (Score:3)
That the human race is largely complete sh*t and they don't deserve the internet in any form, especially the one we envisioned 30 years ago but never realized because the capitalists all moved in and started killing each other over money, squatters rights, scams, spam, and generally unfit behaviour to make a dishonest buck. Professor Farnsworth was right, "I don't want to live!... on this planet anymore!"
Re: (Score:3)
Re: (Score:1)
ClownFlare sucks a big hairy goat dick every morning while getting drilled by the legions of human skin-wearing Reptilians who control the world through the Illuminati. Chairman Sato mines cryptocurrencies using CAPTCHAs.
Now do this to government (Score:1)
Fraud, waste, and incompetence are everywhere. Use AI to find it and fire those responsible. The remaining five government workers will be highly productive!
Re: (Score:3)
Absolutely. Look no further than the White House and the Musk House. It doesn't take AI to identify that level of corruption. Then follow the money to all the companies these guys are *not* cutting off, and you'll find the real "swamp."
It's a start (Score:2)
Which is it today? (Score:3)
Do we love Google for using AI to help them remove some of the cruft, or are we going to continue to hate on AI and say it's worthless?
Re: (Score:3)
Re: (Score:1)
Re: (Score:3)
How do we know the "AI" is correctly identifying fraud? They could be suspending people for false accusations. Google's appeal process is useless.
They didn't solve shit (Score:1)
The real story: 39 Million Fake Accounts on Google (Score:2)