Several views on OpenAI, DeepMind, etc. to open the model to the British government

**OpenAI, DeepMind will open the model to the British government? **

British Prime Minister Rishi Sunak announced Monday at London Tech Week that Google's DeepMind, OpenAI and Anthropic have agreed to open up their artificial intelligence models to the British government for research and security purposes.

British Prime Minister Rishi Sunak said this would allow the UK government** to "help build better assessments to better understand the opportunities and risks of these systems"**. Sunak's speech this time mainly advocates that artificial intelligence is expected to reshape education and medical fields, and heralds the potential of the UK as an "island of innovation".

"* AI is certainly one of the biggest opportunities before us," Sunak said. By combining AI models with the power of quantum, "the possibilities are extraordinary."

In March 2023, the UK government released an artificial intelligence white paper proposing a "Pro-innovation" approach, but recently Sunak has emphasized the need for "Guardrails".

Sunak said on Monday that the UK wants to be not only a core area of scientific knowledge, but also a geographic home for the global regulation of artificial intelligence safety", but the UK has not yet listed specific regulatory proposals and regulations.

Key to these regulations and plans will be the Global AI Safety Summit in the UK this autumn, which Sunak likens to **"the AI version of the UN Conference of the Parties on Climate Change". **

In addition, a working group in the form of a foundation has also been established, currently investing about 100 million pounds in financial support to conduct research on artificial intelligence security technology. The UK first stated that the flexible and balanced regulatory approach in the UK will continue to make it become an attractive investment location.

Anthropic and OpenAI recently established European headquarters in the UK, and Palantir last week announced the opening of an AI research center in the UK.

It is worth noting that OpenAI CEO Sam Altman has been visiting European countries in May 2023, meeting with regulatory agencies of various countries, and meeting heads of government including Spain, Poland, France and the United Kingdom.

Altman first met with Spanish Prime Minister Pedro Sanchez,** because Spain holds the six-month rotating presidency of the European Council this summer, which means that at a critical time in the negotiation of the artificial intelligence rulebook, Spain has a lot to play leeway to influence the discussion. ** And in France, Poland and Spain, regional heads of government are likely to exert influence, through the European Council, on the final form of EU AI norm-setting.

Altman, who also participated in the discussion at UCL, was interviewed on stage, citing ** his preference for regulation that is "somewhere between the traditional European approach and the traditional American approach". **

In May 2023, members of two key committees of the European Parliament, a series of amendments to the risk-based AI regulatory framework originally proposed by the committee (April 2021), aiming to ensure that general artificial intelligence, basic models and generative artificial intelligence Intelligence is outside the scope of the rules.

The European Parliament supports the obligation for providers of underlying models to apply security checks, data governance measures and risk mitigation measures before putting models on the market, which includes a requirement to consider "possibility for health, safety, fundamental rights, the environment, as well as democracy and the rule of law". foreseeable risks".

The amendments will also require base model makers to reduce the energy consumption and resource use of their systems** and to register with EU databases, while providers of generative AI technologies such as OpenAI's ChatGPT will be given transparency obligations It means that AI technology providers must ensure that users are informed that the content is machine-generated and take "adequate safeguards" for content generated by their systems; and provide summaries of any copyrighted material used to train their artificial intelligence. **

However, during this trip to Europe, the CEO of OpenAI took time to meet with British Prime Minister Sunak, and held a meeting with Demis Hassabis of DeepMind and Dario Amodei of Anthropic with Sunak. Of course, the UK remains a major European economy. Also, there are some rumors that OpenAI has been considering setting up a local headquarters in the country. And Anthropic just announced its own London office. For OpenAI, at least in terms of regulation, the UK is no longer a member of the European Union, which means that it cannot participate in the legislation of the EU AI Act.

Beyond that, the Sunak government's attitude is that the government does not intend to create any new domestic legislation to regulate artificial intelligence. Among them, a recent white paper by the British government proposes to rely solely on existing regulatory agencies, such as competition authorities and privacy watchdogs, to provide guidance for the safe development of artificial intelligence, rather than legislating to regulate the use of technology. **

some thoughts

The United Kingdom is likely to benefit from its own geographical advantages and occupy the important field of artificial intelligence security to exert its own influence. **For example, if AI giants support a UK-led dialogue on AI safety research, they can well shape UK AI rules applicable to their own businesses, and the UK will also benefit in this case. **

But if the UK’s AI Security Summit is to be a real and credible success, it must unite researchers around the world and actively engage independent researchers, civil society groups and technical organisations, not just “ Prominent artificial intelligence company" and its own local academics cooperation plan.

**And the global artificial intelligence security summit must abandon prejudice and differences, and seek common welfare for the safety of the entire human being in the AGI world. **