Uber AI lawsuit carries a warning for UK SMEs

Most UK SME owners watching the Uber AI lawsuit unfold will have the same first thought: that is a big-company problem. It is not.

On 2 September 2026, Worker Info Exchange filed a class action at the Amsterdam District Court on behalf of around 240,000 Uber drivers across the UK and Europe. The claim centres on Uber's AI-driven dynamic pay-setting algorithm, which drivers allege breaches GDPR by using automated decision-making and profiling to set personalised fares. The lawsuit also challenges Uber's practice of training its AI models on driver data without adequate consent. This follows a separate €825 million fine from the Dutch data protection authority over automated account suspensions, which Uber says it will appeal. The legal argument rests on rights that apply equally to gig workers and permanent staff. And the compliance rules at the heart of the case apply to every UK employer using AI to influence pay, scheduling, or performance.

What the Uber case is actually about

Strip away the scale and the allegations are straightforward. The drivers say Uber used an algorithm to set pay without telling them how it worked. They say they had no meaningful way to challenge its decisions, and that the AI was trained on their data without a clear legal basis.

Those three failures, lack of transparency, absence of human review, and unlawful data use, are the core of the complaint. None of them require a billion-pound operation to replicate. A fire maintenance firm or small logistics firm using scheduling software that automatically reassigns engineers or shifts. A recruitment agency using AI to score candidate CVs. A sales team manager running commission calculations through an AI tool fed with CRM data. Each of these scenarios can recreate the same legal exposure at a fraction of the cost, and with far fewer resources to defend against a claim.

The Amsterdam claim relies on EU GDPR. The UK equivalent sits in UK GDPR, updated by the Data (Use and Access) Act 2025, which still requires safeguards for solely automated decisions that have a significant effect on a person. "Significant" includes decisions about pay, employment, and performance. Those safeguards include telling people about such decisions and letting them challenge the outcome and ask for human review. Without them, the process is exposed.

Why this matters for HR managers, operations leads, and MDs

For HR managers, the immediate concern is performance monitoring. Many SMEs have adopted AI tools that score attendance, flag productivity dips, or rank employees based on output data. If those scores feed into pay reviews, disciplinary processes, or redundancy decisions without a human reviewing the underlying reasoning, the organisation may already be non-compliant.

For operations managers, the risk sits in scheduling and task allocation. Automated shift planning and AI-driven workload distribution are genuinely useful. They become legally problematic when the system makes binding decisions about a worker's hours or income without any human checkpoint.

For managing directors, the concern is cultural as much as legal. The Uber case, as the drivers describe it, shows a pattern that goes beyond compliance failure: AI used as a control mechanism rather than a capability tool. Drivers had no understanding of how decisions were made, no way to question them, and no sense that the system was designed with their interests in mind. That breeds distrust. In a small business, distrust of management systems is a direct route to talent loss, grievances, and reputational damage. On the drivers' account, the technology worked as designed and the human side failed.

This is the 80/20 reality of AI adoption. The technical implementation rarely causes the crisis. The cultural implementation almost always does.

Three practical steps to take before the end of this month

1. Audit every AI tool that touches staff data. Ask your operations lead or HR manager to list every platform that processes employee information, from scheduling software to performance dashboards to recruitment tools. For each one, establish whether it makes decisions automatically or supports a human who makes decisions. The distinction matters legally.

2. Check your data processing agreements and privacy notices. If you are using an AI tool on staff or customer data, that data use must have a clear legal basis under UK GDPR. Check whether your privacy notice describes the use of automated tools. If it does not, update it. This is a 30-minute task for most SMEs with fewer than 50 staff.

3. Build a human review step into any automated process that affects pay, scheduling, or performance. This does not need to be complex. It means a named person reviews the AI output before any decision is communicated to a worker. Document that review. The ICO will want to see evidence of human oversight if a complaint is made.

None of these steps requires a legal team or a large budget. They require the managing director or HR manager to treat AI governance as an operational discipline, not a legal afterthought.

The honest limitation here

It is worth being direct about what this article cannot do. The Uber case is still in progress. The Amsterdam court has not issued a final ruling. The legal precedent is not yet set. UK GDPR and EU GDPR have diverged since Brexit, so the Amsterdam decision will not bind UK courts directly. That said, UK law still gives people safeguards against significant automated decisions, and the underlying principles of transparency, human oversight, and lawful data use are not in dispute.

The honest caveat for SMEs is this: the legal risk is real, but the timeline of enforcement is uncertain. What is certain is that if an employee or contractor raises a complaint about an automated decision affecting their pay or employment, the absence of documented human oversight will count against you. The cost of a grievance process, an employment tribunal, or an ICO investigation will exceed the cost of getting this right now.

There is also a practical limitation to the three steps above. Auditing AI tools is only useful if someone in your organisation understands what to look for. Many SMEs that have adopted AI tools in the past 18 months have done so without establishing any internal governance habits. The tools work, but the understanding of how they work, and what legal obligations they carry, has not kept pace.

The cultural question MDs need to answer

The Uber case will be studied for years as a governance case study. The legal arguments are important. But the more instructive lesson for a managing director of a 40-person professional services firm or a 150-person logistics operation is the cultural one.

On the drivers' account, Uber treated AI as a mechanism for control. It set pay without explanation, suspended accounts without human review, and trained models on worker data without consent. The workers did not feel like participants in a system designed to help them work better. They felt managed by a machine they could not see or challenge.

AI adoption inside a small business succeeds when staff understand what the tools do, why the business is using them, and what role the tool plays versus the role a manager plays. That is not idealism. It is the practical condition for adoption that sticks, for retained talent, and for a workforce that engages with new tools rather than resisting them.

The organisations that will face legal and retention problems in the next three years are not necessarily the ones using the most powerful AI. They are the ones that deployed AI on their staff without explanation, without consent, and without a plan to build genuine capability alongside the automation.

What to do if you are weighing this up now

gecco's Training and consultancy programme gives SME teams a vendor-neutral grounding in responsible AI adoption, including practical workshops on building assistants with appropriate human oversight and the governance habits that reduce compliance exposure. If your team is adopting AI tools that touch staff or customer data and you want a clear picture of where your governance gaps are, the AI Readiness survey is the right starting point.

Taking the survey gives you access to 65+ free resources, a custom AI Readiness report tailored to your business, and a free 30-minute showcase to walk through the results with a member of the gecco team.

Website · LinkedIn · Case Studies · Newsletter