The legal part…

• With a general election now taking place on 4 July 2024, there’s a question mark over the future of the Data Protection and Digital Information Bill and how data protection law may develop in the next Parliament.
• The Council of the EU gave the final green light to the EU AI Act. The Act will enter into force at some point in June 2024 and most of its provisions will apply 2 years later. We’ll be publishing the first in our series of snapshot articles shortly, to help affected businesses prepare for implementation.
• The government published the key regulators’ strategic approaches to AI. Ministers had requested updates on how the regulators are taking forward the government’s proposals on AI regulation.
• Ahead of the AI Seoul Summit (discussed below) a paper authored by 2 ‘godfathers of AI’ among others warns that current governance plans “fall critically short in view of the rapid progress in AI capabilities”.
• The Information Commissioner’s Office is seeking views until 10 June 2024 on engineering individual rights into generative AI models, in particular in relation to training and fine-tuning. This is the fourth chapter in the regulator’s consultation series on how data protection law applies to the development and use of generative AI.
• The ICO warned organisations developing or using generative AI that they must consider data protection from the outset, as it concluded its investigation into Snap’s ‘My AI’ chatbot. In a recent speech, the Information Commissioner talked about how the ICO applies the law to emerging technologies, including AI.
• The Digital Regulation Cooperation Forum – comprising the ICO, Financial Conduct Authority, Competition and Markets Authority and Ofcom – published the high-level findings from a workshop attended by firms who are deploying generative AI.
• The ICO said that organisations must do more to combat the growing threat of cyber attacks. This follows publication of a report setting out practical advice and case studies for organisations to understand common security failures and take simple steps to improve their security.
• The UK’s consumer connectable product security regime came into force on 29 April 2024, bringing in obligations for businesses involved in the supply chain of these products. See the government’s press release.
• Over in Europe, privacy rights group noyb (‘none of your business’) filed a complaint against OpenAI, the company behind ChatGPT. It argues that ChatGPT provides false information about people which Open AI is unable to correct, in breach of GDPR provisions.
• And the European Parliament approved the Net-Zero industry Act to bolster EU production in technologies needed for decarbonisation.

…and in other news

• The AI Seoul Summit took place on 21 and 22 May 2024 with a focus on AI safety, innovation and inclusivity. Key highlights include: global leaders agreeing to launch the first international network of AI safety institutes; 16 global AI tech companies agreeing new commitments to develop AI safely; 27 countries and the EU committing to deepen work on severe AI risks; and the government unveiling a £8.5 million grants programme to fund research into systemic AI safety.
• An interim international scientific report on the safety of advanced AI informed the discussions. The final report will be published ahead of an ‘AI Action Summit’ to be held in France.
• The UK’s AI Safety Institute recently evaluated 5 large language models already used by the public. Among other things, it found that all of the models are highly vulnerable to basic ‘jailbreaks’ – users attempting to bypass safeguards to elicit potentially harmful outputs – and some will provide harmful outputs even without dedicated attempts to circumvent their safeguards.
• Other work highlighted in the Institute’s fourth progress report include the opening of a San Francisco office and the launch of a new AI safety evaluations platform.
• The finalists of the Manchester Prize – which rewards breakthroughs in AI for the public good – were revealed. The focus this year is on energy, environment and infrastructure.
• The government announced 2 new codes of practice to enhance cyber security in AI and software.
• The National Cyber Security Centre published: guidance on how to disrupt email phishing attacks that target senior executives or budget holders; and joint guidance with the insurance industry to help organisations faced with ransomware demands minimise disruption and the cost of an incident.
• We’ve also seen an update to the NCSC’s machine learning security principles; and a blog post on ransomware attacks and loss of control of data.
• The Digital Standards Initiative of the International Chamber of Commerce launched a complete framework for end-to-end supply chain digitalisation.
• And finally, Meta’s Ray-Ban smart glasses will now post Instagram stories for you without you needing to use a phone, while a Google AI assistant will tell you where you left your glasses.

If you have queries about any of the points covered in this edition of the Technology & Digital round-up, or need further advice or assistance, please get in touch with Sally, Luke or one of our Technology & Digital experts.

More from Walker Morris

Want to watch a previous webinar? Visit our digital academy, home to a library of digital content including webinars, our bite-sized video nuggets and podcasts, including our 60 second videos on what is an NFT and what is a blockchain.

Want to learn more from our Technology & Digital experts and be the first to receive important updates, developments and events from the team? Then visit our #WMTechTalk page or sign up for our newsletter, the Technology & Digital round-up here.