![[Freelance]](../gif/fl3H.png)
What’s been advising government on ‘AI’ policy?
A UK government minister asked for advice on what should be done to encourage use of "artificial intelligence" - advice, that is, from an AI. Freelance journalist Chris Stokel-Walker had the bright notion of making a Freedom of Information request for the ChatGPT logs of Peter Kyle, UK Secretary of State for Science, Innovation and Technology.
A photorealistic image of an artificial intelligence giving advice to a UK Secretary of State for Science, Innovation and Technology on how to encourage use of artificial intelligence, as supplied by ChatGPT's DALL-E feature. It refused to provide an image of Peter Kyle, or of itself, when asked directly.
The ChatGPT logs that Chris eventually received included the minister asking why small and medium businesses (SMBs) have been slow to adopt AI. The answers included: “While the UK government has launched initiatives to encourage AI adoption, many SMBs are unaware of these programs or find them difficult to navigate. Limited access to funding or incentives to de-risk AI investment can also deter adoption.”
The Rt. Hon Peter Kyle MP and flag
On regulatory and ethical concerns the AI said: “Compliance with data protection laws, such as GDPR, can be a significant hurdle. SMBs may worry about legal and ethical issues associated with using AI.”
It's not wrong there.
To its concern about GDPR - the General Data Protection Regulation - we would add issues around copyright and AI and, indeed, the the government's proposals on this. We shall enquire further.
Should the minister have declared pre-emptively that there was a conflict of interest in him seeking advice from ChatGPT on how to expand the use and reach and income potential of ChatGPT?
Chris' initial request for the logs was turned down; it was granted when he refined it to ask only for queries entered “in an official capacity”. The whole thing, as he reports in New Scientist, raises interesting questions about the scope of Freedom of Information requests. On this precedent, a chat with a robot is equivalent to an email exchange with a human, and therefore disclosable.
Revealed: How the UK tech secretary uses ChatGPT for policy advice newscientist.com
Resembling a robot recession did 'AI' crash the markets?
'Artificial Intelligence' our coverage to date
Note A sub-editor writes: the more usual acronym that we have to remember to spell out in policy copy is “small and medium enterprises” (SMEs). It turns out that the UK government - perhaps now to be different from the EU - does refer to “small and medium businesses” (SMBs), so this is as likely to be Kyle's responsibility as ChatGPT's, and we can let it stand. Is there a difference in the policy meaning?