Securing AI from attack
We develop cybersecurity safeguards for AI Models and Systems, where AI is a component in the system that needs defending.
The rapid expansion of Artificial Intelligence (AI) into new industries with new stakeholders – coupled with an evolving threat landscape and huge growth in AI – presents tough challenges for security. ETSI’s Technical Committee on Securing Artificial Intelligence (TC SAI) develops technical standards to combat these challenges. These standards serve as a baseline for ensuring that AI systems are secure, safe, and societally relevant.
Stakeholders impacted by TC SAI activities include all member groups represented in ETSI, as well as broader societal sectors in which AI can be deployed. These encompass end users, manufacturers, operators, and governments.
Key areas of work
Mitigating against AI
We develop countermeasures to mitigate threats from AI systems, where AI is the ‘problem’ (or used to improve and enhance other more conventional attack vectors).
Using AI to enhance security measures against attack
We develop methodologies to strengthen AI security resilience, where AI is part of the ‘solution’
Societal security and safety aspects of the use and application of AI
We develop practical guidance and recommendations for responsible use of AI.
What we do
- Improving the security of AI through developing technical standards that preserve and improve the security of new AI technologies. This includes guidance and evaluation reports to ETSI and its stakeholders on emerging developments in AI.
- Advising and assisting all ETSI Groups on securing AI.
- Ensuring that appropriate standards for securing AI are developed within ETSI.
- Responding to EU policy requests related to securing AI.
Standards & documents
-
ETSI TS 104 033 V1.1.1 (2026-05) Published
Securing Artificial Intelligence (SAI); Security requirements for an Artificial Intelligence Computing Platform
-
ETSI TS 104 158-1 V1.1.1 (2026-03) Published
Securing Artificial Intelligence (SAI); AI Incident Reporting; Part 1: AI Common Incident Expression (AICIE) Global Framework
-
ETSI TS 104 158-2 V1.1.1 (2026-03) Published
Securing Artificial Intelligence (SAI); AI Incident Reporting; Part 2: AI Common Incident Expression (AICIE) Common Container
-
ETSI TR 104 159 V1.1.1 (2026-01) Published
Securing Artificial Intelligence (SAI); Understanding and Preventing Harm from Generative AI
Get involved
For participation inquiries or further information, please contact: [email protected]
Learn more about our membership benefits
Related technologies
Technology
Cybersecurity, privacy and users
Preserving the rights of individuals and organisations to benefit from dependable, cybersecure ICT products and services that are accessible to all
Related news
Press Release
2 June 2026
ETSI releases Standard on Securing AI Computing Platforms
The specification ensures that emergency communications operate successfully despite growing complexity in the network environment.
News
25 March 2026
From vision to action: Strengthening EU–Japan standards cooperation for trusted digital futures
Today, ETSI-TTC Workshop, “How European and Japanese SDOs can support the EU-Japan Digital Partnership”, building on the momentum of ETSI’s mission led by Director General...
Press Release
14 January 2026
ETSI releases world-leading standard for securing AI
Today ETSI announces the publication of its new standard, ETSI EN 304 223, that provides baseline cybersecurity requirements for AI models and systems. Building on...
Press Release
13 November 2025
ETSI publishes new Technical Report on user-centric frameworks for smart, data-driven digital ecosystems
Today, ETSI has published a new Technical Report (TR), which outlines a forward-looking vision for how AI can enable direct user control of personal data....
Press Release
23 April 2025
ETSI Technical Specification sets international benchmark for securing Artificial Intelligence
Today, ETSI unveils its latest technical specification which sets a benchmark for securing Artificial Intelligence (AI) systems against the backdrop of rising cybersecurity threats.
