Are you talking to an AI-powered pro-Israeli superbot? | Interactive news

By the end of 2023, almost half of all internet traffic was made up of bots, a study by US cybersecurity firm found Imperva.

Bad bots reached the highest levels recorded by Imperva, representing 34% of internet traffic, while good bots accounted for the remaining 15%.

This was partly due to the growing popularity of artificial intelligence (AI) for generating text and images.

According to Baydoun, the pro-Israel bots they encountered primarily aim to sow doubt and confusion about a pro-Palestinian narrative, rather than getting social media users to trust them.

Armies of bots – thousands to millions of malicious bots – are used in large-scale disinformation campaigns to influence public opinion.

As bots become more advanced, it becomes more difficult to distinguish between bot and human content.

“AI’s ability to create these larger botnets… has an incredibly detrimental effect on true communication, but also on freedom of expression because they have the ability to drown out human voices,” said Jillian York, director of Freedom of Speech. International Expression of International. non-profit digital rights group Electronic Frontier Foundation.

Evolution of bots

The first bots were very simple, operating according to predefined rules rather than employing the sophisticated AI techniques used today.

In the early to mid-2000s, as social networks like MySpace and Facebook grew, social media bots became popular because they could automate tasks such as quickly adding “friends,” creating user accounts, and automating posts.

These early bots had limited language processing capabilities – they only understood and responded to a narrow range of predefined commands or keywords.

“Previously, online bots, especially in the mid-2010s… were mostly regurgitating the same text over and over again. The text… would obviously be written by a bot,” said Semaan.

In the 2010s, rapid advances in natural language processing (NLP), a branch of AI that allows computers to understand and generate human language, meant bots could do more.

In the 2016 US presidential election between Donald Trump and Hillary Clinton, a study by researchers at the University of Pennsylvania found that a third of pro-Trump tweets and nearly a fifth of pro-Clinton tweets came from bots during the first two debates.

Then came a more advanced type of NLP known as large language models (LLM), which uses billions or trillions of parameters to generate human-like text.