Asked by: Lord Watson of Wyre Forest (Labour - Life peer)
Question to the Department for Science, Innovation & Technology:
To ask His Majesty's Government, further to the Written Answer by Viscount Camrose on 10 April (HL3606), what measures, if any, they intend to implement to ensure accountability for defamatory content generated by large language models.
Answered by Viscount Camrose - Parliamentary Under Secretary of State (Department for Science, Innovation and Technology)
In the white paper consultation response, the Government set out the case for targeted binding measures for the most powerful AI systems to ensure adherence to the principles set out in the AI white paper including safety, transparency and accountability. The Government also set out key questions to consider as we develop our thinking, and we are engaging with stakeholders across academia, civil society, legal experts and regulators on this work. Government has committed to publishing an update by the end of the year.
Asked by: Lord Johnson of Marylebone (Conservative - Life peer)
Question to the Cabinet Office:
To ask His Majesty's Government what plans they have to review their statistical methods regarding the inclusion of international students on postgraduate taught courses in net migration figures.
Answered by Baroness Neville-Rolfe - Minister of State (Cabinet Office)
The information requested falls under the remit of the UK Statistics Authority.
Please see the letter attached from the National Statistician and Chief Executive of the UK Statistics Authority.
The Right Hon. the Lord Johnson of Marylebone
House of Lords
London
SW1A 0PW
19 April 2024
Dear Lord Johnson,
As National Statistician and Chief Executive of the UK Statistics Authority, I am responding to your Parliamentary Question asking what plans there are to review the statistical methods regarding the inclusion of international students on postgraduate taught courses in net migration figures (HL3762).
The Office for National Statistics (ONS) is responsible for publishing long-term international migration statistics. The most recent edition is for year-ending (YE) June 2023[1].
Net international migration accounts for a significant part of population change. Therefore, it is important that long-term international migration estimates account for those joining or leaving the UK population for all reasons, including international students (on both undergraduate and postgraduate courses). This ensures that the size of the population can be measured and the impact of migration on the economy and service requirements can be understood.
However, I recognise that there is also a user need for estimates of specific groups such as international students, as well as the need to understand the contribution of students to net international migration. Therefore, the latest long-term international migration estimates for YE June 2023 include breakdowns by reason for migration.
Alongside the latest estimates, the ONS also published an article named Reason for international migration: international students update: November 2023[2]. This presents further analysis of net migration of non-EU international students up to the YE June 2023 along with cohort analysis that provides insight on the changing behaviours of international students. This includes how long they spend studying in the UK before emigrating or, for example, transitioning onto work visas.
The ONS are continuing to develop their research in this area to provide more granular estimates, further insights on migration patterns of international students and identify priority areas for further research based on user priorities. Afurther update on this work will be published in May 2024, including estimates of international student migration for the YE December 2023.
Yours sincerely,
Professor Sir Ian Diamond
Asked by: Beth Winter (Labour - Cynon Valley)
Question to the Department for Transport:
To ask the Secretary of State for Transport, what his planned timetable is for publishing the review of medical professionals' fees for driver licensing.
Answered by Guy Opperman - Parliamentary Under-Secretary (Department for Transport)
The Driver and Vehicle Licensing Agency (DVLA) is currently engaging with the Professional Fees Committee of the British Medical Association in respect of the fees paid for medical information required for driver licensing purposes.
The DVLA is also reviewing the questionnaires issued to medical professionals when medical information is required to assist in assessing an individual’s fitness to drive; with a view to simplifying the process wherever possible.
The DVLA has introduced a simplified licence renewal process for drivers with epilepsy, multiple sclerosis, some mental health conditions, and glaucoma. This renewal process has significantly reduced the need for the DVLA to seek further information from medical professionals and enabled more licensing decisions to be made based on the information provided by the driver. The DVLA is considering adding more medical conditions to this process.
In July 2022, the law changed to widen the pool of registered healthcare professionals who can provide information as part of the DVLA’s medical investigations into a person’s fitness to drive. Previously, this could only be provided by a doctor. This provides GP surgeries and hospital teams greater flexibility to decide how they manage the DVLA’s requests for information.
These changes will all help to simplify or reduce the number of times the DVLA needs to seek further information from medical professionals.
The length of time taken to process a driving licence application where a medical condition needs to be investigated depends on the condition(s) involved and if further medical information, tests, or examinations are required.
The DVLA, acting on behalf of the Secretary of State for Transport, is required to ensure that all drivers meet the medical standards for driving before a licence is issued. These investigations can often involve seeking further information about an individual’s medical condition(s) from the relevant healthcare professionals. However, it is important to note that the DVLA does not ask medical professionals to assess an individual’s fitness to drive, and as such, figures about how many such assessments have been received from medical professionals are not available.
Asked by: Rachel Hopkins (Labour - Luton South)
Question to the Ministry of Justice:
To ask the Secretary of State for Justice, how many convictions there were for rape in each year since 2005; and what the conviction rate was in the same period.
Answered by Laura Farris - Parliamentary Under Secretary of State (Ministry of Justice) (jointly with Home Office)
This Government is committed to improving the criminal justice system’s response to rape. In 2019, we commissioned our end-to-end Rape Review, publishing this in 2021 alongside a clear Action Plan that committed to delivering sustained improvements for victims. In this Action Plan, we set ourselves stretching ambitions to return the volumes of adult rape cases being referred by the police, charged by the CPS, and going to court back to 2016 levels by the end of this Parliament.
We have exceeded each of these ambitions ahead of schedule. In practice, this means we have more than doubled the volumes of police referrals, charges, and cases reaching court compared to when the Review was commissioned in 2019.
Increases to conviction volumes or conviction rates were not included as ambitions for the Rape Review. The right to a fair trial means that juries should rightly make decisions independently and based on the facts of the case.
Whilst not being a Rape Review ambition, the Ministry of Justice does hold data on convictions for rape offences.
Published data for calendar years (from 2010 to 2022) can be found in the following tool: Outcomes by Offence tool: December 2022.
The most recent published data available, providing data for the years ending in June from 2011 until 2023, can be found in the following tool: Outcomes by Offence tool: June 2023.
The full calendar year for 2023 will be available in the next update of the Outcomes by Offence tool, expected in May 2024.
This can be accessed by navigating to the ‘Prosecutions and convictions’ tab and using the Offence filter to select the following offences in the Outcomes by Offence data tool:
To cover the full period requested, previously unpublished convictions for rape offences from 2005 to 2009 have now been provided in Table 1.
It is not advised to use this data to calculate conviction rate (the number of convictions as a proportion of the number of prosecutions). This is due to the Court Proceedings Database counting two separate records at two separate stages (one for prosecution, one for conviction). We cannot track the defendant throughout their court journey and an individual may appear at each court in separate years, or for a different principal offence at different stages. As a result, this rate is not an accurate measure of the proportion of prosecutions that result in a conviction.
However, the most accessible published data for conviction rates is in the quarterly data summaries for the period 2019/20 onwards published by the CPS in the CPS quarterly data summaries | The Crown Prosecution Service. Furthermore, convictions rates 2007/8 to 2013/14 are available in the CPS Violence against Women and Girls crime report 2013-2014.
Asked by: Angela Eagle (Labour - Wallasey)
Question to the Department for Work and Pensions:
To ask the Secretary of State for Work and Pensions, how many and what proportion of the advance Universal Credit claims flagged by his Department’s machine learning algorithm as potentially fraudulent in which claimants were (a) from protected groups and (b) vulnerable claimants were found to be fraudulent.
Answered by Paul Maynard - Parliamentary Under-Secretary (Department for Work and Pensions)
Publishing these figures would compromise our ability to prevent or detect fraud.
Asked by: Mark Hendrick (Labour (Co-op) - Preston)
Question to the Department of Health and Social Care:
To ask the Secretary of State for Health and Social Care, what her planned timetable is for publishing the revised NHS Accessible Information Standard.
Answered by Maria Caulfield - Parliamentary Under Secretary of State (Department for Business and Trade) (Minister for Women)
NHS England remains committed to publishing the Accessible Information Standard (AIS) as soon as possible, with the intention of publishing it within the first quarter of 2024/25. The standard is within the publication process, and NHS England are working to both progress the documentation through the publication approval process, as well as ensuring that, when published, the AIS and supporting documentation are available in accessible formats.
Asked by: Angela Eagle (Labour - Wallasey)
Question to the Department for Work and Pensions:
To ask the Secretary of State for Work and Pensions, how many and what proportion of the advance Universal Credit claims flagged by his Department’s machine learning algorithm as potentially fraudulent were (a) from protected groups and (b) vulnerable claimants.
Answered by Paul Maynard - Parliamentary Under-Secretary (Department for Work and Pensions)
Publishing these figures would compromise our ability to prevent or detect fraud.
Asked by: Angela Eagle (Labour - Wallasey)
Question to the Department for Work and Pensions:
To ask the Secretary of State for Work and Pensions, how many of the advance Universal Credit claims flagged by his Department’s machine learning algorithm as potentially fraudulent were from (a) protected groups and (b) vulnerable claimants.
Answered by Paul Maynard - Parliamentary Under-Secretary (Department for Work and Pensions)
Publishing these figures would compromise our ability to prevent or detect fraud.
Asked by: Angela Eagle (Labour - Wallasey)
Question to the Department for Work and Pensions:
To ask the Secretary of State for Work and Pensions, how many and what proportion of the advance Universal Credit claims flagged by his Department’s machine learning algorithm as potentially fraudulent were found to be fraudulent.
Answered by Paul Maynard - Parliamentary Under-Secretary (Department for Work and Pensions)
Publishing these figures would compromise our ability to prevent or detect fraud.
Asked by: Angela Eagle (Labour - Wallasey)
Question to the Department for Work and Pensions:
To ask the Secretary of State for Work and Pensions, how many fraudulent advance Universal Credit claims have been identified by his Department's machine learning algorithm.
Answered by Paul Maynard - Parliamentary Under-Secretary (Department for Work and Pensions)
Publishing these figures would compromise our ability to prevent or detect fraud.