Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Cabinet Office:
To ask the Minister for the Cabinet Office, whether he plans to complete an algorithmic transparency report using the Algorithmic Transparency Recording Standard for the Artificial Intelligence Red Box ministers are piloting.
Answered by Alex Burghart - Shadow Chancellor of the Duchy of Lancaster
The government has made transparency around automated decision-making a priority through the publication of the Algorithmic Transparency Recording Standard (ATRS). The ATRS was endorsed by the Data Standards Authority (DSA), and the first approved version was published in January 2023.
The Red Box Copilot is currently in Alpha, being tested in a controlled, transparent manner in a small number of private offices.
Any standards that have been designated by the DSA have an expectation that they will be adopted where there is a use case. As the AI Red Box is being piloted, the Government will consider the publication of an ATRS record, post evaluation.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Cabinet Office:
To ask the Minister for the Cabinet Office, what assessment his Department has made of the effectiveness of the GOV.UK chat experiment.
Answered by Alex Burghart - Shadow Chancellor of the Duchy of Lancaster
Government Digital Service (GDS) tested an early prototype of a Large Language Model (LLM) powered chatbot (GOV.UK Chat) in December 2023. Feedback, and the quality of answers provided, were analysed from 200 users.
Nearly 70% of users found the prototype-generated responses to be useful and the answers were assessed to be accurate 80% of the time.
Ensuring the highest level of accuracy and efficiency is a priority, which is why the next phase of development and testing will focus on how to improve accuracy of the model and explore how users could best discover and use the service.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Department for Science, Innovation & Technology:
To ask the Secretary of State for Science, Innovation and Technology, how many times the National Science and Technology Council met in (a) 2021, (b) 2022 and (c) 2023 as of 23 November.
Answered by Andrew Griffith - Shadow Secretary of State for Business and Trade
It is a long-established precedent that information about the discussions that have taken place in Cabinet and its Committees, and how often they have met, is not normally shared publicly.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Department for Science, Innovation & Technology:
To ask the Secretary of State for Science, Innovation and Technology, pursuant to the Answer of 21 November 2023 to Question 1948 on Artificial Intelligence: Public Sector, and with reference to her Department's policy paper entitled Introducing the AI Safety Institute, published on 2 November 2023, whether it remains her Department's policy to identify new uses for AI in the public sector.
Answered by Saqib Bhatti - Shadow Minister (Education)
As the Department with responsibility for AI policy and strengthening the UK’s capabilities in AI, DSIT will continue to look to identify new uses for AI in the public sector. The work the department started on AI for Good during the summit will support this. We will work with the recently established Incubator for Artificial Intelligence and with departments and public sector organisations leading on identifying use cases in their remit to drive responsible innovation in AI across the public sector.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Home Office:
To ask the Secretary of State for the Home Department, with reference to the Answer of 14 November 2023 to Question 1202 on Members: Correspondence, if he will make an assessment of the status of the technical issue as of 22 November 2023.
Answered by Robert Jenrick
Having reviewed the system we are satisfied that this error is not part of a wider systemic issue.
The majority of all requests for Biometric Residence Permits are successfully processed, however, the case in question was part of a very small number of failures, usually because of a data entry error rather than a system issue. We are taking steps (like postcode lookup and enhanced operational guidance) to further enhance data quality to drive down the small number even further.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Department for Science, Innovation & Technology:
To ask the Secretary of State for Science, Innovation and Technology, with reference to the Answer of 7 November to Question 333 on Social Media: Antisemitism, if she will publish the written responses of the social media companies.
Answered by Saqib Bhatti - Shadow Minister (Education)
The Government agreed to keep written summaries confidential. These summaries contain commercially sensitive information, as well as information about special measures put in place to tackle content related to the conflict, which could be exploited by bad actors if published.
However, the Government has been clear with platforms that they should be adequately protecting their users. The Government is in regular contact with platforms as part of the process of closely monitoring their responses. The Government also convened social media platforms and community groups to facilitate direct feedback about the impact of this content.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Department for Science, Innovation & Technology:
To ask the Secretary of State for Science, Innovation and Technology, pursuant to the Answer of 13 November to Question 334 on Artificial Intelligence: Public Sector, if she will provide a table of all new uses of AI in the public sector her Department has identified as of 14 November 2023; and whether they have been adopted.
Answered by Saqib Bhatti - Shadow Minister (Education)
Each public sector organisation leads on identifying new uses of AI specific to them, and therefore the Department for Science, Innovation and Technology (DSIT) does not identify and track these. DSIT itself has not adopted or deployed any AI tools, it is working to use technologies such as automation and AI innovatively to improve outcomes and efficiency, but we are in the pilot stage at the moment.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Department of Health and Social Care:
To ask the Secretary of State for Health and Social Care, what steps he is taking to implement recommendation 27 of Commercial clinical trials in the UK: the Lord O’Shaughnessy review, published on 25 May 2023.
Answered by Andrew Stephenson
The Government will publish a full response to the Lord O’Shaughnessy independent review into commercial clinical trials in autumn 2023.
The response will include an update on progress and implementation of the initial five headline commitments and foundational actions that the Government made in May 2023 as well as all other recommendations in the review.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Home Office:
To ask the Secretary of State for the Home Department, with reference to the correspondence from the hon. Member for Hove, reference MPAM/1005482/23, what the technical issue is that is delaying his constituent being issued with an accurate biometric residence permit card; and whether this technical issue has affected any other applications made to her Department.
Answered by Robert Jenrick
There has been a technical issue with this case matching the applicant’s biometric information with the production of the Biometric Residence Permit. This incident has now been resolved, and a new BRP will be issued shortly.
Asked by: Peter Kyle (Labour - Hove and Portslade)
Question to the Department for Science, Innovation & Technology:
To ask the Secretary of State for Science, Innovation and Technology, when she plans to publish a response to the consultation entitled A pro-innovation approach to AI regulation.
Answered by Paul Scully
Government is moving quickly to establish the right guardrails for AI to drive responsible, safe innovation. In March, we published the AI Regulation White Paper which set out our first steps towards establishing a regulatory framework for AI.
We will be publishing our response to the consultation later this year, to ensure we can take into account the outcomes of the productive discussions of the AI Safety Summit held at the beginning of November.
Since publishing the White Paper, we have taken rapid steps to implement our approach. We have established a central AI risk function in DSIT which will ensure that we can identify, measure and monitor existing and emerging AI risks. We also announced the world’s first AI Safety Institute which will carefully examine, test and evaluate new types of AI. This builds on our investment of £100 million in the Frontier AI Taskforce – more funding for AI safety than any other country in the world.