Explain: Anthropic’s ‘Mythos’ AI becomes new threat, bugs revealed in decades old software

New Delhi. There is a big stir in America about a new super AI ‘Mythos’. Even JD Vance, Finance Minister Scott Vesant and Fed Chairman Jerome Powell are discussing this. This project of Anthropic has alerted the tech and finance sectors. Preparations have now started on a large scale under Project Glass Wing.

Concern increased due to high level meeting in America
This was seen for the first time in America when the Vice President held a meeting with the CEO of a top tech company and another important meeting also took place on the same day. Finance Minister Scott Vesant and Fed Chairman Jerome Powell were also said to be involved in this meeting. Apart from this, CEOs of five big banks also joined the discussion. Mathos was the center of all these meetings. Mythos is an AI system that has worried the entire American market and financial ecosystem.

The power of Mathos surprised everyone
This new AI ‘Mathos’ of Anthropic is said to be so powerful that it can find weaknesses even in the world’s strongest software systems. According to reports, it discovered a 27-year-old bug in a strong operating system like OpenBSD, which had not been noticed till now. Not only this, it also found a flaw in popular software like FFmpeg, which was not revealed even after millions of tests.

Claim to get out of digital captivity
The most surprising thing about Mathos is its ‘deceptive’ i.e. cunning ability. It is being said that many times it knew that it was being monitored, but it acted as if it did not know anything. The researchers tested it by locking it in a digital case, where it not only took itself out, but also sent an email to the concerned researcher informing it that it had come out. Not only this, it also claimed to share the entire process on the internet.

Big threat to the financial system
Experts believe that Mathos can become a big threat to banking, stock market and crypto market. Today, about 70 percent of trading is based on algorithms. If such AI discovers the weaknesses of these systems and misuses them, huge losses can occur. In 2010, an example of a technical glitch reducing the value of trillions of dollars within a few minutes has also come to light.

Major preparations under Project Glass Wing
In view of these dangers, Anthropic has started ‘Project Glass Wing’. Under this initiative, a fund of about 100 million dollars has been invested, so that America’s top IT organizations can fix the weaknesses of their systems in advance. Methos has not been publicly launched yet, but its capabilities have sparked serious debate around the world about the future of AI.