AI is a big technological advancement and currently plays an important role in many fields and people’s lives. However, this technology can become a machine that produces fake news, spam, data theft… destroys the Internet.
At the end of May, when logging into HBO Max, to confirm “I’m not a robot” users had to solve a series of complex puzzles, such as adding dots to the dice, listening to sounds, etc. the box contains images related to the topic as before. On some other platforms, it also requires users to perform such complex “tasks”. The reason is that the AI has easily passed the normal test, so it takes more effort to prove that it’s not a bot.
The above example is just the tip of the iceberg if you look at how AI is impacting the way the Internet works.
When ChatGPT launched, it ushered in a large-scale AI race. This chatbot is integrated by many companies into their platform to be able to create text and images easily with just one click.
However, when AI prevails on the website, its potential dangers to the Internet have gradually appeared such as fake news, privacy, spam … flooded.
Reddit forum after nearly 20 years of existence has become one of the monuments of the Internet. The special thing of this platform is that it allows third parties to access data for free through APIs to get data to make software and applications.
In April, when ChatGPT got the world’s attention, Reddit changed its mind about a plan to charge companies for accessing its data. Steve Huffman, Director of Reddit affirmed that Reddit’s data warehouse is very valuable in AI training and will not be given away to Google, OpenAI, or Microsoft.
Apollo, which has provided advanced moderation tools to Reddit’s volunteers for decades, also announced the discontinuation of service on the platform. Apollo – Reddit’s decades-old relationship is fractured. One example shows how AI has broken age-old bonds between tech companies.
AI also makes fake news flood the most trusted websites on the Internet. According to NewsGuard, a company that monitors misinformation and assesses the reliability of websites, it has discovered nearly 350 online news sites with almost entirely AI-generated content, including many fake news, Conspiracy theories and hoaxes. These sites have almost no human supervision.
According to a report from Europol, the European Union’s law enforcement agency, it is predicted that in the next few years AI will generate 90% of the content on the Internet.
Although these news sites do not yet have a significant readership, their numbers are growing rapidly. This is a warning sign that the Internet will be flooded with unverified news.
According to Professor Filippo Menczer of Indiana University, there are more and more bots that mimic humans in a sophisticated way and can bypass content censorship barriers to post on X and Facebook platforms.
On e-commerce platforms like Amazon, Etsy also appears a lot of synthetic content from AI. Christopher Cowell from Portland (USA) released a technical textbook, but two weeks ago, a similar one written in AI was available on Amazon. A series of summaries and reviews appeared massively, making users confused, unable to distinguish what is real and what is fake.
Fake news has long been a big problem for the Internet, but AI has contributed to making the situation worse, more out of control, beyond many people’s expectations.
The rise of AI makes security and user privacy more difficult than ever. The introduction of ChatGPT makes online scams harder to detect and target. A non-technical person can use AI to create tools and scam scripts with just a few commands.
Darktrace, a British cybersecurity company, announced that phishing campaigns on the Internet since the beginning of 2023 have increased by 135% compared to the same period last year. By using bots, criminals easily deceive and steal users’ information. AI can even steal passwords from keystrokes.
When community-led initiatives like Reddit and Wikipedia, the Internet is torn apart by artificial intelligence, it feels like it was designed by machines and for machines.
Toby Walsh, an AI professor at the University of New South Wales, thinks that as AI content dominates the Internet, tech companies like Microsoft and Google will have less original data to improve their models. At that time, AI will use its own data to train other AI systems, causing it to degrade and eventually collapse.
John Licato, a professor at the University of South Florida, says the current signs of the web indicate that “the Internet is dying”. When high-traffic websites like Reddit are invaded by bots, humans will no longer be able to create and consume content on the Internet. What is happening now is the harbinger of the death of the Internet.