AI regulation gaps
Need for an AI regulatory framework that effectively addresses the identified challenges of AI governance.
Strongest theme matches
Mixed across source types and ranked by classifier confidence plus text match strength.
Committee recommendation
74match
#18 - 2nd report - Home and travel insurance regulation: more for the FCA to do
We believe that the current regulatory regime is broadly capable of addressing the issues identified during this inquiry, notwithstanding the emerging risks arising from consumers’ growing use of unregulated AI tools. We do not consider that further prescriptive rules are generally the answer. If effectively implemented, supervised and enforced, the Consumer Duty has the potential to address the...
Matched on
terms: regulation
Committee recommendation
70match
#17 - 2nd report - Home and travel insurance regulation: more for the FCA to do
We do not think it is clear whether AI models currently provide consumers with financial advice that would otherwise be regulated if provided by a person or firm. While some witnesses have suggested that the regulatory perimeter needs to be changed, it is not clear that a case for bringing AI tools explicitly inside the regulatory perimeter has...
Matched on
terms: regulation
Committee recommendation
66match
#16 - 2nd report - Home and travel insurance regulation: more for the FCA to do
The growing use of general-purpose AI tools by consumers to research home and travel insurance presents an emerging risk to consumers. They may provide inaccurate or unreliable information and introduce data protection concerns. The FCA considers that these tools currently fall outside the regulatory perimeter and so are not subject to the requirements of the regulatory regime. We...
Matched on
terms: regulation
Committee recommendation
62match
#5 - 2nd report - Home and travel insurance regulation: more for the FCA to do
Consumers are increasingly using AI to research insurance products. While this has the potential to benefit consumers and improve their engagement and understanding of products, the use of AI carries a risk of harm and may provide misleading or incorrect responses.
Matched on
terms: regulation
Committee recommendation
57match
#24 - 5th Report - Data and digital trade
The Government should make every effort to maintain a regular dialogue with counterparts on regulation and legislation on a variety of issues affecting digital trade. These might include regulation of emerging technologies, approaches to data, or digitisation of trade documents. Such cooperation should help ameliorate barriers to trade by working towards providing interoperable regulatory regimes with our closest...
Matched on
terms: regulation
Committee recommendation
57match
#22 - 5th Report - Data and digital trade
The governance of artificial intelligence and its impact on our society and economy is still in its infancy. Regulation of artificial intelligence cannot be undertaken in isolation, and should be considered in cooperation with our global partners.
Matched on
terms: regulation
Committee recommendation
53match
#17 - 5th Report – How to strengthen UK-EU relations
Research cooperation also ensures a deep mutual understanding of our approaches to emerging technologies at an upstream stage, which helps to maintain an overall compatibility of approaches including regulation. (Conclusion, Paragraph 74)
Matched on
terms: regulation
Committee recommendation
53match
#28 - 1st Report - The future of news
Any proposal for a new text and data mining regime must include transparency mechanisms that enable rightsholders to check whether their data has been used. It must offer technical enforceability that goes beyond the likes of robots.txt indicators, which remain inadequate. Meaningful sanctions for non-compliance are essential and the Government’s anticipated IP consultation should explore the options for...
Matched on
classifier match
Committee recommendation
53match
#32 - 5th Report - Data and digital trade
To improve coherence between regulators and departments, we recommend that government departments such as the Department for Science, Innovation and Technology, regulators such as those participating in the Digital Regulation Cooperation Forum, and bodies such as the British Standards Institute and the AI Safety Institute should be included in the multi-stakeholder consultative forum.
Matched on
terms: regulation
Committee recommendation
49match
#41 - 1st Report - Large language models and generative AI
We support the overall White Paper approach. But the pace of delivering the central support functions is inadequate. The regulatory support and co-ordination teams proposed in the March 2023 White Paper underpin its entire success. By the end of November 2023, regulators were unaware of the central function’s status and how it would operate. This slowness reflects prioritisation...
Matched on
classifier match
Committee recommendation
49match
#40 - 1st Report - Large language models and generative AI
Setting the strategic direction for LLMs and developing enforceable, pro-innovation regulatory frameworks at pace should remain the Government’s immediate priority.
Matched on
classifier match
Committee recommendation
49match
#27 - 1st Report - The future of news
While we welcome the new Government’s desire to make progress on this issue, we caution strongly against adopting a flawed opt-out regime comparable to the version operating in the EU. Much better means for ensuring technical viability, transparency, consent and enforcement are needed for a new text and data mining regime to work to UK advantage. If the...
Matched on
classifier match
Committee recommendation
49match
#26 - 1st Report - The future of news
But that does not mean the Government should pursue rules that primarily benefit foreign tech firms (who seem prepared to pay vast sums on energy, computing facilities and staff—but not on data). Previous efforts to find a solution have been weak and ineffectual. The Government must aim for a robust framework that helps the creative industries strike mutually...
Matched on
classifier match
Committee recommendation
49match
#25 - 1st Report - The future of news
The use of news content to train generative AI has the potential to reshape the economics of the media industry. The UK needs a better framework for governing how this works. There are arguments for and against tougher rules. On the one hand, the UK must remain competitive in AI development, or else lose any claim to international...
Matched on
classifier match
Committee recommendation
49match
#20 - 1st Report - The future of news
Advances in generative AI are enabling tech firms to provide engaging and high quality news summaries. This suggests they are increasingly acting as publishers and may need to be regulated as such. Ofcom’s media plurality framework is rapidly becoming outdated, and the previous Government’s years-long timeline for implementing vital changes has been inadequate.
Matched on
classifier match
Committee recommendation
45match
#29 - 4th Report - BBC Royal Charter Review
The challenge of providing accurate, impartial and trusted news to audiences is growing as more people access news through unregulated platforms. It is a particular challenge when it comes to the younger audiences. This trend is exacerbated by the fact that AI is becoming more important in the production and distribution of news. AI can be a tool...
Matched on
classifier match
Committee recommendation
45match
#55 - 1st Report - Large language models and generative AI
The application of the law to LLM processes is complex, but the principles remain clear. The point of copyright is to reward creators for their efforts, prevent others from using works without permission, and incentivise innovation. The current legal framework is failing to ensure these outcomes occur and the Government has a duty to act. It cannot sit...
Matched on
classifier match
Committee recommendation
45match
#44 - 1st Report - Large language models and generative AI
The Government should introduce standardised powers for the main regulators who are expected to lead on AI oversight to ensure they can gather information relating to AI processes and conduct technical, empirical and governance audits. It should also ensure there are meaningful sanctions to provide credible deterrents against egregious wrongdoing.
Matched on
classifier match
Committee recommendation
45match
#42 - 1st Report - Large language models and generative AI
DSIT should prioritise resourcing the teams responsible for regulatory support and co-ordination , and publish an update on staffing and policy progress in response to this report.
Matched on
classifier match
Committee recommendation
45match
#39 - 1st Report - Large language models and generative AI
Extensive primary legislation aimed solely at LLMs is not currently appropriate: the technology is too new, the uncertainties too high and the risk of inadvertently stifling innovation too great. Broader legislation on AI governance may emerge in future, though this was outside the scope of our inquiry.
Matched on
classifier match
Committee recommendation
45match
#35 - 1st Report - Large language models and generative AI
Further clarity on data protection law is needed. The Information Commissioner ’s Office should work with DSIT to provide clear guidance on how data protection law applies to the complexity of LLM processes, including the extent to which individuals can seek redress if a model has already been trained on their data and released.
Matched on
classifier match
Committee recommendation
45match
#30 - 1st Report - Large language models and generative AI
There is a credible security risk from the rapid and uncontrollable proliferation of highly capable openly available models which may be misused or malfunction. Banning them entirely would be disproportionate and likely ineffective. But a concerted effort is needed to monitor and mitigate the cumulative impacts.
Matched on
classifier match
Committee recommendation
45match
#25 - 1st Report - Large language models and generative AI
The Government has made welcome progress on understanding AI risks and catalysing international co-operation. There is however no publicly agreed assessment framework and shared terminology is limited. It is therefore difficult to judge the magnitude of the issues and priorities.
Matched on
classifier match
Committee recommendation
45match
#22 - 1st Report - Large language models and generative AI
We recommend that the Government explores the options for and feasibility of acquiring a sovereign LLM capability. No option is risk free, though commissioning external developers might work best. Any public sector capability would need to be designed to the highest ethical and security standards, in line with the recommendations made in this report.
Matched on
classifier match
Committee recommendation
45match
#20 - 1st Report - Large language models and generative AI
A sovereign UK LLM capability could deliver substantial value if challenges around reliability, ethics, security and interpretability can be resolved. LLMs could in future benefit central departments and public services for example, though it remains too early to consider using LLMs in high-stakes applications such as critical national infrastructure or the legal system.
Matched on
classifier match
Committee recommendation
45match
#13 - 1st Report - Large language models and generative AI
The Government is not striking the right balance between innovation and risk. We appreciate that recent advances have required rapid security evaluations and we commend the AI Safety Summit as a significant achievement. But Government attention is shifting too far towards a narrow view of high-stakes AI safety. On its own, this will not drive the kind of...
Matched on
classifier match
Committee recommendation
45match
#18 - 1st Report - The future of news
More transparency over tech platforms’ algorithms is needed (even if transparency on its own will not be enough). Empowering the regulators quickly is necessary and achievable.
Matched on
classifier match
Scottish FAI
45match
Shomi Moshina and Steven Alexander Wiseman
Consideration should however be given to devising a formal system to ensure so far as can reasonably be done that algorithm outcomes are not overridden save for clear and identified reasons; that the ability of nurse advisors to make that decision should be tested before they are allowed to do so, and should thereafter be regularly reviewed. The...
Matched on
classifier match
Committee recommendation
41match
#38 - 2nd Report - Developing AI capacity and expertise in UK defence
The Government should convene a programme of discussions with the UK’s AUKUS and NATO allies aimed at reaching a mutual understanding of each other’s AI objectives and strategies. These discussions should also explore reaching, where possible, common positions and standards on how AUKUS and NATO allies will develop and deploy AI in defence. These could include common approaches...
Matched on
classifier match
Committee recommendation
41match
#61 - 1st Report - Large language models and generative AI
The IPO code should include a mechanism for rightsholders to check training data. This would provide assurance about the level of compliance with copyright law.
Matched on
classifier match
Committee recommendation
41match
#58 - 1st Report - Large language models and generative AI
If the process remains unresolved by Spring 2024 the Government must set out options and prepare to resolve the dispute definitively, including legislative changes if necessary.
Matched on
classifier match
Committee recommendation
41match
#56 - 1st Report - Large language models and generative AI
In response to this report the Government should publish its view on whether copyright law provides sufficient protections to rightsholders, given recent advances in LLMs. If this identifies major uncertainty the Government should set out options for updating legislation to ensure copyright principles remain future proof and technologically neutral.
Matched on
classifier match
Committee recommendation
41match
#54 - 1st Report - Large language models and generative AI
LLMs may offer immense value to society. But that does not warrant the violation of copyright law or its underpinning principles. We do not believe it is fair for tech firms to use rightsholder data for commercial purposes without permission or compensation, and to gain vast financial rewards in the process. There is compelling evidence that the UK...
Matched on
classifier match
Committee recommendation
41match
#52 - 1st Report - Large language models and generative AI
We urge the Government and regulators to work with partners at pace on developing accredited standards and auditing practices for LLMs (noting that these must not be tick-box exercises). A consistent approach to publishing key information on model cards would also be helpful.
Matched on
classifier match
Committee recommendation
41match
#51 - 1st Report - Large language models and generative AI
Accredited standards and auditing practices are key. They would help catalyse a domestic AI assurance industry, support business clarity and empower regulators.
Matched on
classifier match
Committee recommendation
41match
#49 - 1st Report - Large language models and generative AI
The Government should develop mandatory safety tests for high-risk high-impact models. This must include an expectation that the results will be shared with the Government (and regulators if appropriate), and clearly defined powers to require compliance with safety recommendations, suspend model release, and issue market recall or platform take-down notices in the event of a credible threat to...
Matched on
classifier match
Committee recommendation
41match
#48 - 1st Report - Large language models and generative AI
We welcome the commitments from model developers to engage with the Government on safety. But it would be naïve to believe voluntary agreements will suffice in the long-term as increasingly powerful models proliferate across the world, including in states which already pose a threat to UK security objectives.
Matched on
classifier match
Committee recommendation
41match
#47 - 1st Report - Large language models and generative AI
The Government should ask the Law Commission to review legal liability across the LLM value chain, including open access models. The Government should provide an initial position , and a timeline for establishing further legal clarity, in its White Paper response.
Matched on
classifier match
Committee recommendation
41match
#46 - 1st Report - Large language models and generative AI
Model developers bear some responsibility for the products they are building—particularly given the foreseeable risk of harm from misuse and the limited information available to customers about how the base model works. But how far such liability extends remains unclear.
Matched on
classifier match
Committee recommendation
41match
#45 - 1st Report - Large language models and generative AI
The Government’s central support functions should work with regulators at pace to publish cross-sector guidance on AI issues that fall outside individual sector remits.
Matched on
classifier match
Committee recommendation
41match
#43 - 1st Report - Large language models and generative AI
Relying on existing regulators to ensure good outcomes from AI will only work if they are properly resourced and empowered.
Matched on
classifier match
Committee recommendation
41match
#36 - 1st Report - Large language models and generative AI
The Department for Health and Social Care should work with NHS bodies to ensure future proof data protection provisions are embedded in licensing terms. This would help reassure patients given the possibility of LLM businesses working with NHS data being acquired by overseas corporations.
Matched on
classifier match
Committee recommendation
41match
#34 - 1st Report - Large language models and generative AI
The AI Safety Institute should develop robust techniques to identify and mitigate societal risks. The Government’ s AI risk register should include a range of societal risks, developed in consultation with civil society. DSIT should also use its White Paper response to propose market-oriented measures which incentivise ethical development from the outset, rather than retrospective guardrails. Options include...
Matched on
classifier match
Committee recommendation
41match
#33 - 1st Report - Large language models and generative AI
LLMs may amplify numerous existing societal problems and are particularly prone to discrimination and bias. The economic impetus to use them before adequate guardrails have been developed risks deepening inequality.
Matched on
classifier match
Committee recommendation
41match
#31 - 1st Report - Large language models and generative AI
The AI Safety Institute should develop new ways to identify and track models once released, standardise expectations of documentation, and review the extent to which it is safe for some types of model to publish the underlying software code, weights and training data.
Matched on
classifier match
Committee recommendation
41match
#29 - 1st Report - Large language models and generative AI
The AI Safety Institute should publish an assessment of engineering pathways to catastrophic risk and warning indicators as an immediate priority. It should then set out plans for developing scalable mitigations. (We set out recommendations on powers and take-down requirements in Chapter 7). The Institute should further set out options for encouraging developers to build systems that are...
Matched on
classifier match
Committee recommendation
41match
#28 - 1st Report - Large language models and generative AI
There are however no warning indicators for a rapid and uncontrollable escalation of capabilities resulting in catastrophic risk. There is no cause for panic, but the implications of this intelligence blind spot deserve sober consideration.
Matched on
classifier match
Committee recommendation
41match
#26 - 1st Report - Large language models and generative AI
The Government should publish an AI risk taxonomy and risk register. It would be helpful for this to be aligned with the National Security Risk Assessment.
Matched on
classifier match
Committee recommendation
41match
#10 - 1st Report - Large language models and generative AI
Large language models have significant potential to benefit the economy and society if they are developed and deployed responsibly. The UK must not lose out on these opportunities.
Matched on
classifier match
Committee recommendation
41match
#1 - 1st Report - Large language models and generative AI
Large language models (LLMs) will have impacts comparable to the invention of the internet.
Matched on
classifier match