Google doesn't want to make its AI algorithm public

Trillium Asset Management, a shareholder who owns $135 million worth of shares in Alphabet (Google's parent company), has just made a request for transparency in AI (artificial intelligence) and algorithms at its 2023 annual meeting. The fund representative said that algorithms can lead to dangerous results in some areas such as justice and medicine, and that accountability and transparency are necessary "to know whether this technology is safe for society," Trillium argued.

This isn’t the first time the foundation has asked Alphabet for transparency. Trillium made similar requests before OpenAI’s ChatGPT AI exploded in early 2023. Google declined both times.

Google keeps some secrets about algorithms and AI to protect against bad guys

According to the search giant, information about the algorithm is public on the website, along with transparency about how YouTube organizes suggested content for users. But the proprietary algorithms that are the basis of the company's business operations cannot be published.

"Algorithmic transparency also needs to consider the risks when information is exploited by bad actors, affecting user privacy and revealing sensitive commercial information," Google explained.

Mr. Sundar Pichai - CEO of Google commented that AI will have an impact on all areas of life, including healthcare , and affirmed that the company is approaching artificial intelligence "responsibly" to make this technology more useful.

The company's announcement and the request from the Trillium fund come amid the rapid development of AI, along with concerns from users and the technology industry. Last May, Geoffrey Hinton, a pioneering AI engineer, left Google and warned about the dangers of AI-based chatbots.

Source link: https://thanhnien.vn/google-khong-muon-cong-khai-thuat-toan-ai-185230605090700037.htm