WHO GOVERNS AUTONOMOUS PUBLIC TRANSPORT?

The Emerging Governance Gap Inside Britain’s Autonomous Bus Trials

The debate around autonomous and driverless public transport is often presented as a technological story.

The focus is usually placed on:

- artificial intelligence,

- sensors,

- automation,

- innovation,

- digital infrastructure,

- and the future passenger experience.

But after months of Freedom of Information work examining the wider system of bus safety governance across Britain, a much bigger and more important question is beginning to emerge beneath the technology itself:

«Who actually governs the safety of autonomous public transport systems once they begin operating on public roads?»

Because while Britain is now moving steadily towards autonomous passenger transport trials and deployment, the governance structures behind those systems appear far less visible, far less coherent, and far more fragmented than many people might reasonably assume.

And that matters enormously.

Not simply because autonomous buses are technologically complex, but because they represent something fundamentally different within the transport system:

- software-driven decision-making,

- public-road AI deployment,

- shared operational responsibility,

- distributed accountability,

- and hybrid public/private governance structures operating inside publicly used transport networks.

The issue is therefore no longer simply:

«“Can the technology work?”»

The issue increasingly becomes:

«“How is this being governed, monitored, overseen and held publicly accountable?”»

And based on disclosures recently obtained through Environmental Information Regulations requests relating to autonomous bus trials in Greater Cambridge, that question is becoming increasingly difficult to answer clearly.

---

THE CAMBRIDGE AUTONOMOUS BUS DISCLOSURES

The disclosures relate to the “Connector” autonomous bus projects operating in the Greater Cambridge area.

The requests sought information relating to:

- safety governance frameworks,

- safety management plans,

- incident reporting procedures,

- accident investigation responsibilities,

- safety risk oversight,

- governance structures,

- liability arrangements,

- performance monitoring,

- safety data publication,

- and comparative oversight arrangements between autonomous and conventional bus operations.

At first glance, the responses repeatedly attempt to establish a very clear position:

«Cambridgeshire County Council does not hold most of the core safety governance information.»

Throughout the response, responsibility for key safety documentation is instead repeatedly attributed to what the authority describes as the “Commercial Entities”:

- Fusion Processing,

- Alexander Dennis,

- Whippet Coaches,

- and associated delivery partners.

The authority states that:

- safety management documentation is held by the commercial entities,

- incident reporting arrangements are held by the commercial entities,

- safety cases are held by the commercial entities,

- risk registers are held by the commercial entities,

- operational monitoring is held by the commercial entities,

- governance documentation is held by the commercial entities,

- and liability frameworks are held by the commercial entities.

The repeated phrase appearing throughout the disclosure is:

«“The Council does not hold this information.”»

Yet when the response is read carefully alongside the accompanying quarterly review and project governance slides disclosed under the same request, a much more complicated picture begins to emerge.

---

THE GOVERNANCE STRUCTURES CLEARLY EXIST

Because despite the repeated positioning that the authority does not hold the wider safety picture, the disclosed material simultaneously confirms the existence of formal governance and safety oversight structures operating around the trials themselves.

The disclosed documents reference:

- Project Safety Committees,

- Safety Committee Reports,

- Independent Safety Reviews,

- structured quarterly governance reviews,

- operational sign-off procedures,

- risk reporting structures,

- and formal project board oversight arrangements.

The governance structures become particularly visible within extracts disclosed from a grant offer agreement connected to the project.

Those extracts confirm the existence of:

- a formally established “Project Safety Committee,”

- responsibilities to identify and minimise hazards,

- monitoring of corrective actions,

- formal reporting into Project Board structures,

- and governance provisions preventing safety decisions from being overridden without agreement from the Project Safety Committee.

The disclosed clauses are highly significant.

They state:

«“Project Safety Committee: A committee established by Alexander Dennis and Fusion to identify and minimize potential hazards and safety risks during the Project, and to monitor the implementation of all corrective actions...”»

The documentation further confirms:

«“the Project Board shall not be entitled to make any decision which in the Project Safety Committee’s reasonable opinion affects or concerns matters of safety...”»

This is not an absence of governance.

It is evidence of a layered governance structure operating around autonomous passenger transport deployment.

And that is where the central public-interest issue begins to emerge.

---

THE CENTRAL GOVERNANCE QUESTION

If public authorities are:

- participating in project governance,

- hosting quarterly review mechanisms,

- receiving safety committee reporting,

- facilitating deployment on public roads,

- integrating autonomous systems into wider transport strategy,

- and acting as accountable bodies for government-backed funding streams,

then to what extent can those same authorities reasonably position themselves as not holding the wider safety governance picture?

Because the disclosures appear to reveal a system where:

- governance exists,

- safety oversight structures exist,

- operational reporting exists,

- risk management exists,

- safety sign-off exists,

- project boards exist,

yet responsibility for holding the “full” safety picture appears fragmented across multiple commercial and institutional actors.

That fragmentation becomes even more significant because autonomous public transport systems do not operate in isolation.

They operate:

- on public highways,

- among pedestrians,

- cyclists,

- conventional traffic,

- vulnerable road users,

- and within publicly integrated transport systems.

And unlike many conventional transport technologies, autonomous systems fundamentally depend upon:

- software behaviour,

- algorithmic interpretation,

- machine learning,

- sensor systems,

- disengagement procedures,

- operational intervention frameworks,

- and shared technical responsibility between multiple organisations.

Which means governance itself becomes part of the safety system.

---

A FRAGMENTED ACCOUNTABILITY STRUCTURE

One of the most striking aspects of the Cambridge disclosures is the apparent distribution of responsibility across numerous different actors.

The material suggests:

- operators hold some information,

- technology providers hold some information,

- regulators hold some information,

- police and DVSA would become involved post-incident,

- commercial entities hold core safety cases,

- public authorities host partnership structures,

- government departments fund elements of the programme,

- while public-facing governance visibility remains comparatively limited.

This creates an obvious question:

«Who ultimately holds the complete operational safety picture once autonomous systems are operating within publicly integrated transport networks?»

Because at present, the governance structure appears heavily distributed.

And that pattern mirrors wider themes now repeatedly emerging across broader bus safety governance work nationally:

- fragmented oversight,

- distributed accountability,

- inconsistent transparency,

- unclear ownership,

- operational reform moving faster than governance maturity,

- and public-interest information frequently sitting behind commercial structures.

---

THE “NONE” RESPONSES MAY BE THE MOST IMPORTANT PART

Some of the most revealing answers within the disclosure are actually the shortest.

In response to requests for:

- commitments to publish safety performance data,

- governance comparisons between autonomous and conventional bus oversight,

- incident reporting held by the partnership,

- or governance comparison reports,

the response repeatedly states:

«“None.”»

That matters enormously.

Because autonomous public transport is not a niche experimental curiosity anymore.

It is increasingly becoming:

- transport policy,

- public safety policy,

- governance policy,

- AI oversight policy,

- public accountability policy,

- and future infrastructure policy.

Yet the disclosures suggest that the systems surrounding transparency and public visibility may still be immature, fragmented or underdeveloped compared with the scale of the technological transition now underway.

---

THE WIDER NATIONAL CONTEXT

The Cambridge disclosures do not exist in isolation.

They sit alongside a much wider pattern already emerging through broader Freedom of Information work into bus safety governance across Britain.

Across Combined Authorities, franchising bodies, regulators and devolved transport structures, recurring themes continue to emerge:

- fragmented governance,

- unclear accountability,

- inconsistent safety reporting,

- variable transparency,

- distributed oversight structures,

- and differing levels of public visibility.

In Greater Manchester:

- safety publication commitments have been delayed,

- commercial sensitivity arguments have been raised,

- and governance responsibility has appeared split between GMCA and TfGM.

In emerging devolved systems:

- governance structures are still “being developed,”

- safety transparency frameworks are often not yet fully visible,

- and responsibilities continue transferring between institutions while oversight arrangements evolve.

In other areas:

- authorities state they do not hold safety data,

- or distinguish between operational data and governance accountability,

- while reform, franchising and integration continue accelerating.

And now autonomous transport appears to be reproducing many of the same governance tensions:

- distributed responsibility,

- unclear ownership,

- fragmented accountability,

- and uncertainty around who holds the complete picture.

---

TECHNOLOGY IS MOVING FASTER THAN GOVERNANCE

This may ultimately be the most important issue of all.

Because the autonomous bus debate is often framed around technological readiness.

But governance readiness may prove equally important.

Britain may be developing:

- increasingly sophisticated automation systems,

- AI-assisted mobility,

- integrated transport technologies,

- digital optimisation systems,

- and autonomous passenger operations,

while still lacking:

- mature public accountability structures,

- consistent transparency frameworks,

- unified safety oversight models,

- and clearly visible governance arrangements capable of bringing the entire operational safety picture together coherently.

That creates an obvious long-term public-interest concern.

Because trust in autonomous public transport will not depend solely upon whether the technology functions.

It will depend upon whether the governance surrounding the technology is:

- understandable,

- transparent,

- publicly accountable,

- independently scrutinised,

- and visibly capable of responding when things go wrong.

---

THE LESSON FROM RAIL AND AVIATION

Britain already understands this principle elsewhere in transport.

Rail has:

- RAIB,

- structured reporting systems,

- independent investigation,

- and formalised governance visibility.

Aviation has:

- AAIB,

- highly developed safety cultures,

- mandatory occurrence reporting,

- and deeply embedded independent oversight structures.

Yet bus safety governance remains comparatively fragmented.

And autonomous bus governance may now be exposing just how underdeveloped wider governance structures still are within road passenger transport generally.

Because once software systems begin participating directly in operational decision-making within public passenger networks, the traditional distinctions between:

- operator,

- regulator,

- technology provider,

- authority,

- and infrastructure manager

become increasingly blurred.

Which means governance clarity becomes even more important, not less.

---

THE FUTURE QUESTION

The Cambridge autonomous bus disclosures therefore raise a much bigger question than simply:

«“How safe are autonomous buses?”»

The bigger question increasingly becomes:

«“Is Britain building governance systems capable of safely governing autonomous public transport at scale?”»

Because the issue is no longer just technological capability.

It is governance capability.

And based on the evidence emerging so far, that governance framework may still be evolving far more slowly than the technology itself.

---

ABOUT THE AUTHOR

Lee Odams is:

- RMT Branch Secretary of the Nottinghamshire & Derbyshire Bus Branch

- Secretary of the RMT National Industrial Organising Conference of Bus Workers

- an independent transport governance and safety transparency campaigner

His ongoing Freedom of Information work examines:

- bus safety governance,

- franchising accountability,

- safety transparency,

- public transport oversight,

- devolution,

- and emerging transport governance structures across Britain.

---

TAGS

#AutonomousVehicles

#AutonomousBuses

#PublicTransport

#TransportSafety

#BusSafety

#RoadSafety

#TransportGovernance

#ArtificialIntelligence

#FutureMobility

#ConnectedVehicles

#TransportPolicy

#PublicAccountability

#SafetyGovernance

#FreedomOfInformation

#BusIndustry

#MobilityInnovation

#AI

#ConnectedTransport

#SmartMobility

#TransportTransparency