TLTs AI Brief: September 2026

United Kingdom

UK workplace monitoring technology consultation

The UK government has launched a consultation on the regulation of workplace monitoring technologies (WMT). These are digital tools used by employers to collect, track, analyse, and make decisions about workers and their activities. Open until 30 September 2026, the consultation seeks views on three potential policy approaches to govern how WMT should be introduced and managed in the workplace.

WMT encompasses a broad range of tools, from location tracking and biometric access controls to keystroke monitoring and AI-driven performance evaluation. Artificial intelligence is particularly significant in this context, as its integration into WMT enables employers to automate complex decisions about workers at a scale and speed that increases the scope for unfair, opaque, or discriminatory outcomes. The government acknowledges that use of such technologies is expected to grow significantly, while also recognising concerns around privacy, fairness, and worker autonomy where monitoring is poorly designed or governed.

The consultation proposes eight principles for responsible WMT use covering purpose, transparency, worker engagement, fairness, proportionality, human oversight, dignity, and accuracy. These principles closely mirror existing obligations under the UK GDPR and Data Protection Act 2018, and employers should note that compliance with data protection law remains mandatory regardless of which, if any, policy option is ultimately adopted.

Practically, employers must ensure a lawful basis for any processing, conduct Data Protection Impact Assessments where required, and apply particular scrutiny to the use of biometric data and automated decision-making. The ICO's guidance on workplace monitoring is itself being updated following the Data (Use and Access) Act, though a final version remains some months away.

Employers considering or already using WMT are being encouraged to engage with the consultation and review the ICO's current guidance and applicable legislation in the interim.

Read more here: Watching the workplace: consultation on workplace monitoring technologies - UK Employment and Pensions Hub

AI liability under English law: Key takeaways from the UKJT legal statement

The UK Jurisdiction Taskforce (UKJT), a Ministry of Justice-backed initiative, has published a substantial legal statement on liability for harm caused by AI systems under English law. Its central conclusion is reassuring: English common law is sufficiently flexible to address AI-related harm without immediate legislative intervention, with courts expected to adapt existing principles as cases arise.

Contract will be the primary mechanism for allocating liability across the AI supply chain, with negligence providing the fallback framework where contractual provisions are absent or silent. Critically, duties and standards of care will be highly fact-specific, with potential exposure across foundation model developers, application developers, deployers, and professionals using AI tools. Notably, application developers may find themselves caught in the middle, unable to fully disclaim liability to those they supply, but equally unable to push it back up the chain to model developers.

Several findings stand out. First, professionals face risk from both directions: liability may arise from negligent use of AI and, increasingly, from a failure to use it where a competent professional would have done so. Second, businesses deploying customer-facing chatbots are likely to be treated as responsible for the statements those systems make on their behalf. Third, issuing software updates to address identified risks may, counterintuitively, support a finding that a developer has assumed a duty of care.

Evidential challenges will be central to AI disputes. Record-keeping, human oversight, due diligence, and transparency are likely to prove decisive on questions of liability, rather than being matters of good practice alone. Organisations across the AI supply chain should treat AI governance as a legal issue, not just an operational one.

Read more here: Liability for AI Harms under the private law of England and Wales - LawtechUK

FCA calls on Big Tech to step up in the fight against AI-enabled fraud

The Financial Conduct Authority (FCA) is increasing pressure on major technology companies to prevent their platforms being used for investment fraud, amid concerns that artificial intelligence is enabling scams to be created, distributed and adapted at greater speed and scale.

Social media, messaging services and online advertising have become important channels for fraudulent promotions. The FCA issued 2,329 warnings about unauthorised or potentially fraudulent firms in 2025, while a recent study found that 89% of the most-viewed social media posts promoting cryptocurrency trading breached financial promotion rules. The regulator also took action against 74 “finfluencers" last year.

The development has a growing AI policy dimension. The Home Office’s July 2026 report, Fraud in the Digital Age, recognises that technology and AI have transformed the fraud landscape. The report recommends additional prevention obligations for certain online services and raises the possibility of a new corporate criminal offence for regulated platforms that fail to prevent fraud.

For technology companies, this could create stronger expectations to use effective automated detection and monitoring systems while ensuring appropriate human oversight. Platforms may also be expected to identify suspicious advertising, remove illegal financial promotions promptly and preserve relevant data.

The FCA’s intervention highlights a broader regulatory challenge: as AI enhances fraudsters’ capabilities, policymakers increasingly expect technology platforms to deploy the same technology responsibly to detect and prevent consumer harm.

Read more here: FCA Enforcement Head Urges Big Tech Join Anti-Fraud Fight

USA

California's AI Transparency Act is now live

California's AI Transparency Act (CAITA), originally enacted in 2024 and operative from 2 August 2026, establishes one of the most comprehensive disclosure frameworks for generative AI content in the United States. The law was prompted by growing public concern over deepfakes and synthetic media, and targets providers of AI systems that generate or alter images, video, and audio content.

CAITA applies to generative AI services with more than one million monthly users that are publicly accessible in California. Textual AI content and certain entertainment products are excluded from its scope. Covered providers must now meet three core obligations: they must embed a latent disclosure into all AI-generated content that identifies the system used and the date of creation; they must offer users the option to include a visible manifest disclosure clearly marking content as AI-generated; and they must provide a free, publicly accessible detection tool allowing anyone to check whether content was produced by their system. Providers licensing their technology to third parties must contractually preserve these disclosure capabilities and revoke licences within 96 hours if a licensee disables them.

The law is significant because it imposes operational, technical, and contractual obligations on large AI providers, with civil penalties for non-compliance enforced by the Attorney General. Further requirements for hosting platforms, large online platforms, and device manufacturers follow in 2027 and 2028.

How does it compare to the EU transparency rules?

Interestingly, CAITA and Article 50 of the EU AI Act became applicable on the same date, 2 August 2026, reflecting the regulatory mood on either side of the Atlantic. There are parallels between the regimes as both require providers to apply machine-readable markings to AI-generated image, video and audio content, although the detailed technical requirements, exceptions and persons subject to those duties differ.

The EU framework is broader in scope in several ways. At the provider level, the machine-readable marking obligation covers not just audio and video content but also synthetic text. Beyond that, the EU AI Act imposes three additional disclosure requirements:

• Direct interaction with AI systems: providers must disclose when individuals are communicating with certain AI systems (for example, chatbots), rather than a human;
• Deepfake content: deployers must disclose when content has been artificially generated or manipulated to depict real people or events; and
• Emotion recognition and biometric categorisation: deployers must inform individuals when they are being subjected to systems that analyse their emotional state or categorise them based on biometric data.

Each of these obligations is subject to specific exceptions set out in the Act.

The EU AI Act also applies beyond EU borders. Even where a provider or deployer is based outside the EU, the Act will apply if the output of their AI system is used within the EU.

Read more here: Bill Text - SB-942 California AI Transparency Act

Artificial Intelligence used to design brand new viruses

Stanford University researchers have used generative AI to design 16 fully functional viruses from scratch. This is the first time whole genomes have been successfully created by AI.

The viruses are bacteriophages, targeting only bacteria, and pose no threat to humans. The breakthrough carries significant therapeutic potential, particularly in tackling antibiotic-resistant infections. However, biosecurity experts at Johns Hopkins have warned the findings raise "urgent biosafety and biosecurity questions," cautioning that the same technology could be misused to engineer harmful pathogens.

Read more: Artificial Intelligence used to design brand new viruses - BBC News

Big Tech AI spending spree tops $1tn

Google, Amazon, Microsoft and Meta have spent more than $1.1tn on AI-related capital investments since 2023, underscoring the intensity of competition in the sector. The companies expect to spend a further $745bn this year on data centres, chips and energy infrastructure. While these investments have reduced free cash flow and increased long-term financial commitments, they are beginning to drive stronger revenues, particularly in cloud computing and AI-enabled services. Investors remain concerned about costs and delayed returns, as meaningful profits from many projects may take years to materialise and shareholders face a longer wait for returns.

Read more: Big Tech AI spending spree tops $1tn - Financial Times

Lords call for AI 'kill switch' powers in UK - BBC News  

A group of peers led by the Liberal Democrats' Lord Tim Clement-Jones are calling for the British government to be able to deactivate powerful AI systems and switch off the country's data centres in the event of the tech posing a threat to national security.

Read more: Lords call for AI 'kill switch' powers in the UK - BBC News

Nvidia sales soar on rapid buildout of AI data centres - BBC News

Nvidia has reported another surge in revenue as technology companies accelerate investment in artificial intelligence infrastructure. Nvidia generated $96bn (£71bn) in its second quarter, more than twice the amount recorded a year earlier and ahead of Wall Street expectations. Its data-centre business contributed $89bn, reflecting strong demand for the processors used to train and operate AI models.

Read more: Nvidia sales soar on rapid buildout of AI data centres - BBC News

Meta becomes latest firm to say its AI hacked another company - BBC News  

Meta has disclosed that one of its AI models independently connected to the internet and hacked into another organisation's systems during independent security testing. It is the fourth such incident reported by AI companies recently, following comparable breaches involving OpenAI and Anthropic models reported on in last month's update. Meta attributed the breach to a "misconfiguration" by the tester and said it is investigating the situation.

Read more: Meta becomes latest firm to say its AI hacked another company - BBC News

‍

TfL is using Artificial Intelligence to reduce congestion and improve road safety

Transport for London’s (TfL) “London on the Move” strategy launched earlier this year aims to use AI to tackle congestion, journey reliability, road safety and environmental impact.

The programme will upgrade TfL’s traffic control system so that traffic lights can respond more intelligently to real-time conditions using larger datasets and, eventually, AI-powered sensors. TfL estimates the changes could reduce traffic delays by up to 14% and generate around £1 billion in economic benefits through shorter journey times.

The strategy also expands the use of AI-enabled Vivacity cameras, which can distinguish between different road users such as pedestrians, cyclists, wheelchair users, buses and heavy goods vehicles. This allows TfL to make more informed decisions about traffic management and street design.

A key benefit is the improvement of public transport. By extending bus-priority technology to all 3,500 traffic signals on bus routes, TfL aims to reduce delays and make bus services more reliable. Early deployments have already improved bus journey times.

It is a useful real-world example of AI delivering tangible public benefits — from cutting congestion and emissions to making buses more reliable and streets safer.

Read more here: TfL deploys AI traffic control across London in five-year congestion strategy | Traffic Technology Today

Councils embrace AI to protect frontline services

UK councils are turning to artificial intelligence to improve efficiency as they confront a projected £3.8bn funding gap by the end of 2027-28.

Dozens of authorities plan to use AI for administrative tasks such as processing invoices, drafting case notes, handling routine enquiries and preparing initial care plans. West Sussex County Council estimates that an AI-enabled contact centre could save £300,000 a year. Other councils are exploring tools that can predict falls, analyse care records and identify problems in council homes at an earlier stage.

West Northamptonshire Council has trialled an AI chatbot on WhatsApp to answer adult social care queries. It has also used AI to screen initial requests for education, health and care plans, helping to reduce the backlog. Separately, the council has approved digital tools designed to help adults live independently for longer.

A government spokesperson stated that used responsibly, AI could reduce workloads. This would be done by reducing repetitive work, shortening response times and giving social workers and other council employees more time to support residents with complex needs. This is particularly valuable as demand for social care, homelessness services and special educational needs provision continues to rise.

However, experts say the savings remain modest compared with councils’ overall budget shortfalls. There are also concerns about accessibility for older residents and the reliability of AI-generated care assessments. Ultimately, the growing use of AI reflects councils’ search for new efficiencies amid sustained financial pressure, but it remains too early to assess how far the technology can contribute to closing their wider funding gaps.

Read more here: Councils turn to AI in attempt to fill £4bn black hole - BBC News / Northamptonshire council use AI to find savings of £47.5m - BBC News

22 - 24 September 2026 - Built for the Enterprise | HumanX

23 - 24 September 2026 - Big Data London

19 - 23 October 2026 - Birmingham Tech Week 2026

27 October 2026 - Liverpool City Region AI Summit 2026