Meta's AI has begun to improve itself

In his policy report on the future of artificial intelligence, the tech company founder wrote that general AI smarter than humans, one of the key measures of superintelligence, is on the horizon.

Over the past few months, we've begun to see signs that our AI systems are improving themselves.

Progress is slow for now, but it is undeniable.

According to Independent Turkish, one of the main fears about superintelligence, frequently voiced by academics and industry leaders, is that a self-improving system could lead to an AI that is out of control and rogue.

While Zuckerberg did not address these concerns, he warned that advanced AI carries serious risks.

We believe the benefits of superintelligence should be shared with the world as widely as possible. However, superintelligence will also raise new security concerns.

We will need to be diligent in mitigating these risks and be careful about what we open source. However, we believe that building a free society requires us to aim to empower people as much as possible.

He claimed the firm's approach to superintelligence differs from its rivals because it focuses on empowering individuals rather than automating "all valuable work."

Noting that billions of people worldwide use Meta products such as Facebook, Instagram and WhatsApp, the billionaire said his company is in the best position to integrate artificial intelligence into people's daily lives.

If successful, he noted, it would result in people "spending less time on productivity software and more time creating and connecting."

The note came just ahead of the company's quarterly earnings release, which saw better-than-expected financial results following a major investment in artificial intelligence.

Earnings of $7.14 per share on revenue of $47.52 billion beat Wall Street expectations of $5.92 per share and revenue of $44.8 billion.

Meta's share price rose by double digits as investors appeared hopeful that the tech giant's multi-billion dollar gamble on artificial intelligence development would pay off.

"In some ways, this race for superintelligence is similar historically to the race for the personal computer, the internet browser, the search engine and the smartphone," Mike Proulx, research director at Forrester, told The Independent.

But the main difference is that this race goes much faster because the AI that Meta and the others are racing against can accelerate itself.