Greece presents AI to transcribe and check GSM-R rail dialogues

AI system

System: Speech-to-text and AI protocol-compliance monitoring of GSM-R stationmaster-driver communications (supplier not named)
Model: Not assessed
Release: Not assessed
Application developer: Not named
Role: Not assessed
Configuration: Not assessed
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: The proposed system transcribes conversations over GSM-R railway radio and checks them against communication protocols. Control-center staff review any departures from those protocols. [1]
Human oversight: Not assessed
Relation to AI: Kyranakis described speech transcription and checks against communication protocols. The original interview does not establish that the system was operating routinely or that an alert prevented an incident. [1]
AI contribution: Not assessed

Outcome identifiers: Not assessed
Benefit: Greece’s transport minister described AI transcription of railway radio conversations and checks against communication protocols. He gave no measured safety result. [1]
Category: Increased Product Safety
Status: Not assessed
Beneficiary groups: Organizations. Society. People.
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Participants: Developer: Not named. Operator: Hellenic Ministry of Infrastructure and Transport (planned).
Jurisdictions: Operator: Greece.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Conditions for replication: The system needs to transcribe Greek railway speech accurately. Control-center staff need to review and act on the conversations that it flags. [1]
Replication attempts: Not assessed
Later findings: The retained interview describes the system’s intended function. It establishes no later operational result. [1]
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Source numbers: AIBD-0257-source-0001 [1]
Citations: Numbered references below. [1]
Language: Not assessed
Source relationships: Independent [1]
Original evidence: The broadcaster’s written interview reports the minister’s description. It gives no measured safety result or confirmation of routine use. The broadcast itself was not inspected. [1]
Access and versions: SKAI’s complete written report is available, dated 18 May 2026 at 09:03. [1]
Supported findings: Not assessed
Evidence links: Not assessed
Evidence strength: Not assessed
Missing evidence: Not assessed

Assessed benefits: Organizations. Society. People.
Depth: Not assessed
Reasons: Organizations: No outcome is documented. Society: No outcome is documented. People: No outcome is documented.
Observed or intended: Not assessed
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: Independently reported

Kind: Not assessed
Related entries: Not assessed
Field: Railway safety. Operational communications compliance monitoring.
Purpose: The system is intended to monitor operational radio communications and flag departures from protocol for the control center.
Dates: Source published: 18 May 2026. Updated: 24 October 2026. [1]
Stage: Not assessed
Version: SHA-256: a90659c74483…

News report Independent author

SKAI, “Κυρανάκης στον ΣΚΑΪ: Σύστημα τεχνητής νοημοσύνης θα ελέγχει σε πραγματικό χρόνο τις συνομιλίες στα τρένα,” May 18, 2026, 09:03; report of its interview with Deputy Minister Konstantinos Kyranakis.

Original journalism: SKAI reported its own interview with the responsible minister. The functional and safety claims remain ministerial testimony, not an independent technical evaluation.