Shulou(Shulou.com)06/03 Report--

2020-01-08 12:59:23

Lai Ke is from au Fei Temple.

Qubit report | official account QbitAI

The export to China is restricted on this side, and the domestic regulation is relaxed there.

Feeling the anxiety of staying ahead, the United States has recently released 10 principles in order to promote the development of AI.

The US chief technology officer said the set of principles was intended to remove barriers to private sector AI innovation and growth.

Ten principles: promoting AI Innovation and growth

These ten principles are important on the agenda of the Pelant government following the signing of the American AI Initiative in February 2019.

It is stated in the principle document:

Promoting AI growth and innovation is a high priority for the United States government.

Federal agencies must avoid acts that unnecessarily hamper AI innovation and growth in terms of regulatory or unregulated behavior.

Risk assessment and cost-benefit analysis are conducted before any regulatory action is taken against AI, focusing on a flexible framework rather than one-size-fits-all regulation.

The specific contents of the ten principles are as follows:

1. Public trust in AI, government regulatory and unregulated approaches to AI must promote reliable, robust and reliable AI applications.

two。 In public participation, institutions should provide "ample opportunity" for the public to participate in all stages of the rulemaking process.

3. With regard to the integrity of science and the quality of information, institutions should seek verifiable evidence in an open and objective way to obtain technical information about AI, which can not only provide information for policy decisions, but also enhance public trust in AI.

4. In risk assessment and management, risk-based methods should be used to determine which risks are acceptable, which may bring unacceptable harm, or where the expected cost is greater than the expected benefit.

5. Benefits and costs, institutions should carefully consider all social benefits and distribution effects before considering regulations.

6. Flexibility, regulations should be adjusted to adapt to the rapid changes and updates of AI applications.

7. Fairness and non-discrimination, institutions should consider the issues of fairness and non-discrimination in the results and decisions resulting from artificial intelligence applications.

8. Openness and transparency, transparency and openness can enhance public trust and confidence in AI applications.

9. Safety and security, institutions should pay special attention to existing controls to ensure the confidentiality, integrity and availability of processes and information stored and transmitted by artificial intelligence systems.

10. Inter-agency coordination, agencies should coordinate with each other to share experiences and ensure the consistency and predictability of AI-related policies to promote innovation, growth, and AI in the United States.

American AI ambition

Eric Schmidt, former CEO of Google and its parent company, and Robert Work, a former deputy secretary of defense, told ABC in November that the role of the United States as the supreme authority on AI research could be in jeopardy and that competitors like China were on the rise and had strong momentum in some key areas.

With regard to previous bans on "face recognition" by some state and local governments, government officials said that AI was overregulated.

The Trump administration is reluctant to participate in the Global artificial Intelligence Cooperation Program, working with seven other countries to develop principles and choose to take the lead in publishing their own principles. It is even hoped that the "American Standard" will become an important core of global standards.

Kratsios, the White House technology adviser, says this standard is crucial when authoritarian competitors such as China and Russia pursue their aggressive AI agendas.

An administration official said that the White House's AI principle will serve as an example for other Western countries.

Judging from the attitude of the EU in advocating the AI Code of Ethics last year, it is clear that the Trump administration is not on the same path as his "imaginary partner".

But the White House still wants the EU to be in the same camp as itself:

Europe and our partners should avoid using strict innovation stifling models. The best way to oppose the dictatorship's use of AI is to ensure that the United States and our international partners are global centers of innovation.

The relevant experts in the United States are not optimistic about going their own way.

Martijin Rasser, a senior researcher at the New Center for Security, said: if the US insists on acting alone, its influence may be small.

This is worrying, and it is disadvantageous for us to take a different path from other countries.

Document transfer gate

Https://www.whitehouse.gov/wp-content/uploads/2020/01/Draft-OMB-Memo-on-Regulation-of-AI-1-7-19.pdf

Reference link:

Https://www.theverge.com/2020/1/7/21054653/america-us-ai-regulation-principles-federal-agencies-ostp-principles

Https://www.wired.com/story/white-house-favors-light-touch-regulating-ai/

Https://federalnewsnetwork.com/artificial-intelligence/2020/01/white-house-releases-first-of-its-kind-set-of-binding-ai-principles-for-agency-regulators/

Https://www.toutiao.com/i6779432202535961091/

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.