Shulou(Shulou.com)11/24 Report--

April 26 (Xinhua) the high cost of developing artificial intelligence technology has led many startups to rely on the cloud platforms of large technology companies such as Microsoft or Amazon to develop their products. AI Now, a research group, says this will only make big technology companies stronger. At present, there is an urgent need to solve this imbalance immediately.

AI Now, a research group founded in 2017 by Tuyuan Pexels, released a report last week detailing the impact of large technology companies on the development of artificial intelligence, arguing that the development of artificial intelligence "fundamentally depends on" resources such as data and computing power controlled by large technology companies.

Many theories about the development of artificial intelligence come from large technology companies, such as the fact that artificial intelligence needs unrestricted innovation to benefit society, and that the development of artificial intelligence is related to social progress, the report said. AI Now believes that this statement is good for large companies and even affects the views of many policy makers on technology.

"if we want to make any meaningful changes in the technology industry, we need to start by facing the concentration of influence," says Sarah Myers West, managing director of AI Now.

For the research and development of artificial intelligence, the data model supporting the whole system requires intensive development and huge computing power to play a role.

The good news is that the tools, resources and computing power to develop these data models are readily available, which has led to the recent explosion of artificial intelligence, such as OpenAI's artificial intelligence chat robot ChatGPT and Runway's text and video editors.

But the bad news is that the infrastructure that supports these artificial intelligence and huge computing power are concentrated in the hands of a small number of big companies, and this prosperity will only contribute to their influence. Almost all imaginable artificial intelligence tools rely on cloud services, such as Amazon AWS, Microsoft Azure or Google Cloud, or some combination of the three.

In fact, even if artificial intelligence startups compete with these technology companies, they are contributing to the profits of these big technology companies. The more times startups use major cloud platforms, the more fees they pay to competitors.

This suggests that while people have been worried about the monopolies of large technology companies, the rise of artificial intelligence will only make them stronger.

For the research and development of artificial intelligence, the data model supporting the whole system requires intensive development and huge computing power to play a role.

A serious problem with the imbalance of competition in artificial intelligence is that large technology companies are competing head-on with these smaller start-ups. The difference is that large technology companies have access to resources and data that are not available to outsiders. West calls this concept "gatekeeper control".

For example, Microsoft has warned Azure cloud service customers that they cannot use data from Bing search engine to train artificial intelligence. But Microsoft and its artificial intelligence partner OpenAI are using the data to develop new products.

There is fierce competition in the artificial intelligence market. A few months after ChatGPT became popular, OpenAI released the latest version of GPT's flagship model to strengthen its lead. This has forced Google to hastily release its own artificial intelligence technology to stop the rise of Microsoft's integration of artificial intelligence new Bing.

West says there is a way to encourage competition and reduce the reliance on large technology companies for artificial intelligence technology development.

"if there is one thing we have learned from the struggle against big technology companies over the past decade, it is that we cannot let companies lead this dialogue," she said. "it is up to regulators and the public to define the future of artificial intelligence."

AI Now has made several policy recommendations to prevent large technology companies from further strengthening their established position. One of these is the concept of data minimization, which allows companies to collect only the necessary data, not large amounts of data. Large technology companies usually have the most data, and in most cases they have the computing power to take advantage of that data.

Better enforcement of antitrust laws and linking competition to the concept of privacy could limit the size of large technology companies, says Mr West. Once all this is in place, it will provide a more level playing field for other companies.

"We need to achieve this by strengthening competition enforcement, and the good news is that competition regulators are paying attention to this issue," said Mr West. "

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.