Shulou(Shulou.com)11/24 Report--

141listed companies advertise on AI-generated junk websites.

And you may not know it?

A new analysis shows that there were as many as 393 ads on 55 AI-generated sites between May and June this year.

The manufacturers involved include six major banks and financial services companies, four luxury department stores, three well-known sports brands, and so on.

An example of △, and these "grab" advertising sites are likely to be unregulated.

For example, on an employment website identified as generating information from AI, the ChatGPT language appears in the title of an article:

Sorry, as an artificial intelligence language model, I can't access external links or websites by myself.

△ source: NewsGuard, that is to say, these AI sites not only create useless spam on the Internet, but also steal advertising fees from ordinary websites, and even the financiers and dads are kept in the dark.

What's going on?

Google is also the NewsGuard for "accomplices" to report such information, and they have been tracking AI Unreliable Artificial Intelligence-Generated News websites (UAIN for short) recently.

In the past month, the number of such sites they have observed has increased from 49 to 217.

Many of them use AI to publish a large number of articles, even more than 1200 a day. But there are few humans responsible for editing and supervision.

For example, some articles are AI adapted from some media messages, the same pictures and wording will appear, and the article may also be interspersed with the characteristic language of AI: "as an artificial intelligence language model, I can't provide information about XXX."...

The purpose of the apology body of △ ChatGPT is to create advertising space for itself.

But the question is, such an obvious junk website, why can you still receive advertisements?

This starts with Google's advertising mechanism.

NewsGuard found that of the 393 ads placed on AI, 356 were distributed by Google ads.

This delivery is based on algorithms to determine the website, audience, and so on. Under normal circumstances, you can block some junk websites when launching, but from the current situation, it is likely that the AI site is growing so fast that Google has not had time to ban it.

This has violated Google Advertising's own rules: ads cannot be placed on automatically generated spam pages.

NewsGuard sent four emails to Google about the situation, asking Google to comment on how AI sites make money from advertising. Google gave an initial reply and asked for more details, but there was no updated reply as of June 25.

AI is killing the traditional Internet, but this is just a sign that AI is messing with the Internet.

In fact, many major websites have been hit to varying degrees by AI in the past few months.

Websites such as Reddit, Wikipedia and Stack Overflow all began to change under the influence of AI.

Before that, there was a heated discussion that Reddit wanted to increase the cost of visiting API, and the company believed that the high-quality data on the platform should not be whacked by AI. As a result, this was met with dissatisfaction from users, and hundreds of communities boycotted it.

Wikipedia may be numb to this kind of whoring. Now they want to talk more about how to use AI to enrich their own content. After all, AI has great advantages in terms of knowledge coverage and generation speed, but it is difficult to guarantee the authenticity and quality of the content, which still needs to be verified by human beings.

Stack Overflow was the first website to issue a temporary ban on ChatGPT.

The reason is that there is too much junk information produced by AI at this stage, and human beings do not have a good control method for the time being.

But the change of the Internet is already happening quietly.

The senior author of The Verge said:

The impact of AI at this stage is only a process in the long-term development of the Internet. In essence, this is a war about information-who makes it, how to get it, and who gets paid.

Now a new Internet is being born, and the choices we make now will determine how it develops.

Reference link:

[1] https://www.newsguardtech.com/misinformation-monitor/june-2023/

[2] https://www.theverge.com/2023/6/26/23773914/ai-large-language-models-data-scraping-generation-remaking-web

This article comes from the official account of Wechat: quantum bit (ID:QbitAI), by Mingmin.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.