TM Forum flags AI trust gap in telecom

TM Forum says telecom operators are moving fast on AI, but most cannot prove their systems are trustworthy. The gap could slow the industry’s push toward autonomous networks as regulators tighten AI rules starting in 2026.

Why it matters: - TM Forum says trust is becoming a make-or-break issue for communication service providers as they move toward autonomous networks, composable IT and AI-driven operations. - The research suggests many operators may be unable to prove their AI is safe, secure and reliable at scale, even as they increase deployment. - Regulators in Europe are also raising the bar, adding pressure to show evidence of AI governance, auditability and sovereignty.

What happened: - TM Forum CEO Nik Willetts opened DTW Ignite 2026 in Copenhagen with a warning about a growing AI trust gap in telecom. - New research from TM Forum Insights, in partnership with IBM's Institute for Business Value, found 72% of communication service providers believe their AI is trustworthy. - Only 14% of the 130 AI decision-makers surveyed across 130 operators worldwide said they can produce evidence to prove that trust. - TM Forum launched the Race to 2030 strategy at DTW Ignite to help CSPs become AI-native businesses built on autonomous networks, composable IT and trusted AI.

The details: - The report, Why trust and assurance are key to AI success, says many operators are advancing toward autonomous flows across networks, IT, data and ecosystems without the operational proof needed to validate those flows. - Willetts said trustworthiness is the top driver of telco brand value worldwide, ahead of coverage and price. - He added: “Get trust right, anything is possible. Get trust wrong, and nothing else matters.” - TM Forum says the industry needs evidence-based assurance to prove autonomous networks are safe, composable architectures are secure and AI-led decisions are dependable. - The EU AI Act will require high-risk AI systems to meet audit and governance standards from August 2026. - The proposed Cloud and AI Development Act would add enforceable sovereignty assurance levels for public sector cloud and AI procurement. - TM Forum says the research indicates regulation can speed readiness by increasing operational pressure while improving AI risk posture for organizations that take governance seriously.

Between the lines: - The research points to a familiar enterprise problem: confidence in AI is rising faster than the controls needed to verify performance, safety and compliance. - For telecom operators, that gap could become more costly as they shift from pilots and governance decks to production AI in core network and IT systems. - TM Forum is positioning trust, assurance and sovereignty as the foundation for the next phase of AI adoption, not just compliance add-ons.

What's next: - DTW Ignite 2026 will feature a Trustworthy AI and Data Summit focused on the move from AI ambition to operational proof. - In the Mission Garage, delegates will see live demos, Catalyst projects and sessions on responsible AI, AI orchestration and assurance, agentic interaction security, MODaaS, data products and AI-native delivery. - TM Forum says the goal is to help CSPs build production-ready AI that is auditable, interoperable and trusted at scale. - The full report and a related leadership discussion paper on sovereign AI are available from TM Forum. - More information is available in the full report.