OpenAI sued over school shooting as UK weighs AI safety and datacentre growth

AI under legal fire: the case that could reshape tech accountability

The lawsuit filed by British Columbia against OpenAI and its CEO Sam Altman marks a turning point in the debate over artificial intelligence and public safety. The Canadian province alleges that the February mass shooting at a school in Tumbler Ridge could have been prevented if the company had alerted local law enforcement to the shooter’s use of ChatGPT to plan the attack. The case, brought in a San Francisco federal court, seeks damages for recovery efforts and demands changes to how OpenAI monitors conversations that may lead to violence.

This legal action arrives as the UK grapples with its own regulatory framework for AI. While the British government has positioned itself as a leader in AI safety—hosting global summits and proposing guardrails for high-risk applications—the Tumbler Ridge case exposes a critical gap: the absence of clear protocols for tech companies to report potential threats identified in user interactions. The lawsuit does not accuse OpenAI of direct liability for the shooting, but it challenges the industry’s long-held immunity under Section 230 of the US Communications Decency Act, which shields platforms from legal responsibility for third-party content.

For UK policymakers, the case underscores the urgency of defining what constitutes a "duty to warn" in the age of generative AI. The Online Safety Act, which came into force last year, imposes obligations on platforms to mitigate harm, but it does not address the specific risks posed by AI-driven planning tools. As the UK’s Competition and Markets Authority (CMA) prepares to release its final report on AI foundation models next month, the Tumbler Ridge lawsuit may force a reckoning with whether existing regulations are fit for purpose—or if new legislation is needed to hold companies accountable for the unintended consequences of their technology.

Datacentres: the unseen engine of AI’s growth

While legal battles dominate headlines, the physical infrastructure powering AI is expanding at an unprecedented scale. Alibaba Cloud’s announcement this week of a plan to reach 20 gigawatts (GW) of datacentre capacity by 2032—nearly double its current footprint—highlights the intensifying global race for computational power. The Chinese tech giant’s CEO, Eddie Wu, framed the expansion as a necessary step to support what he called "Machine Thinking," a vision of AI as a ubiquitous force akin to electricity or steam engines.

The UK, home to one of Europe’s largest datacentre markets, is feeling the strain of this growth. Last year, the National Grid warned that datacentres could consume up to 20% of the country’s electricity by 2030, prompting calls for stricter planning controls. Alibaba’s plans include a new custom chip designed to improve energy efficiency, but critics argue that such innovations merely delay the inevitable: a collision between AI’s insatiable demand for power and the UK’s net-zero commitments.

The tension is particularly acute in London and the Southeast, where local councils have begun rejecting datacentre applications on grounds of energy consumption and visual impact. Morphotonics, a Dutch deeptech firm that raised €40 million this week to develop next-generation display technology, exemplifies the dilemma. While its innovations could reduce the energy footprint of datacentres, the company’s expansion into European markets—including the UK—will depend on securing reliable power supplies, a challenge that has already led some firms to consider relocating to countries with greener grids.

The human cost of digital infrastructure

Beyond the legal and technical debates, the week’s developments lay bare the human consequences of unchecked technological expansion. The severed Verizon fibre cable in New Jersey, which disrupted hundreds of flights across the US Northeast, serves as a stark reminder of how fragile digital infrastructure remains. The incident, caused by construction crews, led to ground stops at all three major New York City airports and exposed the lack of redundancy in critical systems. While Verizon denied responsibility, the Federal Aviation Administration (FAA) has launched an investigation into whether air traffic control networks are sufficiently resilient to such failures.

In the UK, the rollout of biometric age verification for online services—introduced last year to protect minors—has reignited concerns about privacy and surveillance. A recent report by the Information Commissioner’s Office (ICO) found that nearly half of young people now trust AI over humans to verify facts, a statistic that underscores both the growing reliance on automated systems and the erosion of critical thinking skills. The trend is particularly pronounced among Gen Z, who have come of age in an era where algorithms curate news, entertainment, and even social interactions.

Yet even as AI permeates daily life, its environmental toll is becoming harder to ignore. The divers who spent six hours clearing waste from a river in Northampton this week—part of a broader effort to address pollution from datacentre cooling systems—offer a tangible example of the hidden costs of digital growth. As the UK’s datacentre capacity expands, so too will the demand for water and energy, raising questions about whether the benefits of AI justify its ecological footprint.

What to watch

The British Columbia lawsuit against OpenAI is unlikely to be resolved quickly, but its implications for UK regulation will be immediate. The CMA’s forthcoming report on AI foundation models is expected to recommend stricter oversight of high-risk applications, including mandatory reporting of harmful user interactions. Meanwhile, the UK’s Department for Science, Innovation and Technology (DSIT) is under pressure to clarify whether the Online Safety Act applies to AI-generated content, a question that has taken on new urgency in the wake of the Tumbler Ridge case.

On the infrastructure front, Alibaba’s 20GW target will test the limits of the UK’s energy grid. The company’s new chip, designed to reduce power consumption by up to 30%, may offer a temporary reprieve, but long-term solutions will require collaboration between tech firms, regulators, and energy providers. The UK’s decision to delay its ban on new petrol and diesel cars until 2035—citing concerns about grid capacity—suggests that datacentres could face similar regulatory hurdles in the years ahead.

For consumers, the week’s events serve as a reminder that innovation is rarely neutral. The rise of "Big Toothbrush," with its £600 smart brushes and prebiotic toothpastes, reflects a broader trend: the commodification of everyday health, where convenience often comes at the expense of accessibility. As AI and digital infrastructure reshape industries, the challenge for policymakers will be to ensure that progress does not deepen existing inequalities—or create new ones.