One proposal to preserve the artificial intelligence companies in the test: Insurance

Rapid height for AI Tolide It has built almost billions of dollars in industry, but the members of the committee in the southwest of the largest artificial intelligence companies are facing less supervision than companies lining up outside the Austin Conference, Texas.

“There is more organization on a more dining truck than this technology,” said Emilia Javorsky, director of the Future Program at the Future of Life Institute, during a committee in SXSW on Monday.

“Part of this is that we understand the dining truck more than we understand this technology,” Daniel Kokotolo, a former safety researcher at the Shatti Obay maker who left the company in 2024 to talk about the need for additional safety and transparency in the industry.

The active height of artificial intelligence and technology complexity makes it particularly difficult to formulate rules and inserting political considerations to accomplish something.

But there are a lot of ideas, including one of the Harvard Lawrence Lesij University Law Professor: Claiming the developers of artificial intelligence to carry insurance.

“It is really important to understand the regulations that are not destroy, it can, if it is smart,” Lesij said on the stage in the plate.

The need for transparency

The idea of ​​”trust” was in the minds of many speakers about artificial intelligence in SXSW, with experts not only about the potential for science fiction for separators experts, but also about the daily practical effects of bias and errors in more chatting products such as Chatbots.

The question is how to make sure that every Amnesty International developer builds that confidence. The integrity of artificial intelligence is the subject of companies that companies know that they need to be addressed. Just watch the pages allocated to them in Openaiand Googleand Microsoft, manand Adobe Elsewhere.

Javorsky proposed an irreplaceable organizational structure about the development of prescribed medications: Think of the Food and Drug Administration of Amnesty International. Most of the resources used in the development of drugs are to prove that they are safe, which creates confidence in these products as soon as they reach the market. When consumers feel a safe product, they are less worried about using it.

“Even from the perspective of a person who just wants to see artificial intelligence flourishes as an industry, safety is something that you should care about,” said Javorsky.

He paid Cocotglo and Leese to protect those who are violations of workers in artificial intelligence companies. The laws of those whose violations are not entitled to workers who advocate illegal activity, but many activities that are likely to be dangerous include new technologies such as legal artificial intelligence. There are no laws on books for these types of things.

Lesij the European Union said Artificial intelligence regulations Very complicated and do not force companies to do the right thing by default.

“I think instead we must develop a serious conversation about the minimum viable products for organization,” he said. “The viable product is to build incentives inside these companies to act in the right way.”

Insurance as the list of Amnesty International

Lesij said that the demand of artificial intelligence developers to carry insurance will create the market incentives for responsibility. Insurance companies will specify prices based on the risk of company products and behaviors, which means that it will be one of the most expensive to produce products that lack caffeine handrails.

The requirements of insurance coverage are found in other contexts to put a price, mainly, on risky and irresponsible behavior. Think about how to increase car insurance rates if you have a weak driving record and bottom if you have a safe driving date.

“It will be raw in the beginning, and then you will have some insurance companies that come out and think about how to market its price differently,” Lesij told CNET after the painting. “This is the type of motivation that we need to create this type of competitive effect to determine exactly the safety risks they need to ensure.”

Lesyj said that the insurance condition will not resolve everything, but it will create some incentives for companies to develop strong and accountable safety requirements because they pay the most intelligent and more intelligent artificial intelligence systems.

“The insurance, for me, is the thing that addresses the reckless risks that this will lead to severe damage to society,” he said.