The UK government wants to make a hard axis to boost its economy and industry with AI, and as part of it, it rotates an institution that founded a year ago for a very different purpose. Today the Department of Science, Industry and Technology has announced that it will rename the AI Security Institute at the “AI Security Institute”. (The same first letters: Same URL.) With this, the body will shift from the main exploration of areas such as existential risk and bias into large linguistic models, in the center of cyberspace, in particular “Strengthening the protection against the dangers posed by AE in national security and crime”.
At the same time, the government has also announced a new cooperation with humanity. No fixed services have been announced, but the MOU suggests that the two will “explore” using Anthropic Claude’s AI assistant in public services. And the man will seek to contribute to work in scientific research and economic modeling. And at the AI Security Institute, it will provide tools to evaluate the AI’s capabilities in the context of safety risks.
“AI has the ability to transform how governments serve their citizens,” said flower co -founder and CEO Dario Amodei in a statement. “We look forward to exploring the way AI’s AI assistant could help UK government services to strengthen public services, with the aim of discovering new ways to make vital information and services more efficient and affordable to residents of the United Kingdom “.
Humanity is the only one announced today – which coincides with a week of AI activities in Munich and Paris – but is not the only one working with the government. A series of new tools revealed in January were all the Openai fed. (At that time, Peter Kyle, Minister of Technology, said the government was planning to work with various fundamental AI companies and this is evidenced by humanity.)
The alternation of the government from the government’s Security Institute-which started just a year ago with many fanfares-in AI’s security should not come as an excessive surprise.
When the recently established work government announced AI’s plan to change in January, it was noteworthy that the words “security”, “damage”, “existential” and “threat” did not appear in the document at all.
This was not supervision. The government’s plan is to start investing in a more modernized economy, using technology and in particular AI to do so. She wants to work closely with Big Tech and also wants to build her own big techniques.
Helping this, the main messages they promote were growth, AI and greater growth. Civil servants will have their own AI assistant called “Humphrey”, and are encouraged to share data and use AI in other areas to accelerate their way of working. Consumers will get digital wallets for their government documents and chatbots.
So have AI security issues resolved? Not exactly, but the message seems to be that they cannot be examined at the expense of progress.
The government claimed that despite the change of name, the song would remain the same.
“The changes I announce today represent the logical next step in the way we approach AI’s responsible development – helping us release AI and develop the economy as part of our plan for change,” Kyle said in a statement. “The work of the AI Security Institute will not change, but this renewed focus will ensure that our citizens – and those of our allies – are protected by those who seem to use AI against our institutions, democratic values and our way of life. “
“The focus of the Institute from the beginning was on security and we have created a team of scientists focusing on the evaluation of serious risks to the public,” added Ian Hogarth, who remains president of the Institute. “Our new team of malignancy and deepening cooperation with the National Security community mark the next stage of dealing with these risks.”
Further, priorities appear to have changed around the importance of “AI Security”. The biggest risk that the AI Security Institute is currently considering in the US is that it is going to be disassembled. US Vice President JD Vance telegraphed much earlier this week during his speech in Paris.
TechCrunch has an AI -focused newsletter! Sign up here to get it to your inbox every Wednesday.
