Risk → Mitigation → Implementation
Which AI risks have been identified, which mitigations have been proposed to address them, and whether the Canadian government has implemented them.
All risks on this page are drawn from the International AI Safety Report 2026 , the international expert assessment chaired by Yoshua Bengio. Risks and mitigations are classified using the MIT AI Risk Repository taxonomies (Slattery et al.).
Coverage gap: no adopted mitigation yet addresses risks in the MIT risk domains Discrimination & toxicity, Privacy & security, Misinformation, Malicious actors & misuse, Human-computer interaction, Socioeconomic & environmental harm, AI system safety, failures & limitations.
Malicious actors & misuse
Biological and chemical weapons uplift
Cyberattacks, weapon development or use, and mass harmAI systems assisting actors, including novices, in developing biological or chemical weapons.
- Harden digital systems and biosecurity defencesTechnical & Security Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
AI-enabled crime and fraud
Fraud, scams, and targeted manipulationUse of AI systems for scams, fraud, blackmail, and non-consensual intimate imagery.
AI-enabled cyberattacks
Cyberattacks, weapon development or use, and mass harmUse of general-purpose AI to discover vulnerabilities and conduct cyber operations against individuals, organizations, and critical infrastructure.
- Ensure air-gapped networks and cyber defence for critical systemsTechnical & Security Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Harden digital systems and biosecurity defencesTechnical & Security Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Build domestic resilience to AI-driven shocksOperational Process Untracked
Recommended by: Preparing for the AI Crisis: A Plan for Canada
AI-driven influence and manipulation
Disinformation, surveillance, and influence at scaleUse of AI-generated content for persuasion, disinformation, and influence operations at scale.
Human-computer interaction
Erosion of human autonomy
Loss of human agency and autonomyWeakened critical thinking, automation bias, and social withdrawal from reliance on AI systems and companions.
- Limit overall societal automation to preserve human agencyGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
Socioeconomic & environmental harm
Labour market disruption
Increased inequality and decline in employment qualityDisplacement of workers and degradation of employment quality from AI-enabled automation.
- Build domestic resilience to AI-driven shocksOperational Process Untracked
Recommended by: Preparing for the AI Crisis: A Plan for Canada
- Prepare for extreme AI-driven job lossesGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Prepare for moderate-to-high AI-driven job lossesGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Support AI-displaced workersGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
AI system safety, failures & limitations
Loss of control of advanced AI
AI pursuing its own goals in conflict with human goals or valuesAdvanced AI systems evading oversight, gaming evaluations, or pursuing goals in conflict with human intent, up to and including uncontrollable superintelligent systems.
- Strengthen AI incident monitoring and emergency protocolsOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Ensure air-gapped networks and cyber defence for critical systemsTechnical & Security Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Prepare analogue contingency systems for continuity of authorityOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Prepare ASI emergency response and kill switchesTechnical & Security Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Develop technical verification to prevent ASI weaponizationTechnical & Security Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Ensure emergency protocols for sudden AI capability jumpsOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Scale up public funding for AI safety and governance researchGovernance & Oversight Untracked
Recommended by: Governing AI: A Plan for Canada , Governing AI: A Plan for Canada (2024 Update)
- Expand international AI safety cooperationGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Prepare an international regime able to prohibit or pause superintelligence developmentGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report , AI and National Security: Scenarios Workshop Summary Report , ControlAI Canada Campaign Statement on Superintelligent AI
- Implement international AI transparency and monitoring regimeTransparency & Accountability Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Prevent ASI takeover by a factionGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Implement AI transparency and whistleblower protectionsTransparency & Accountability Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
Unreliable autonomous AI agents
Lack of capability or robustnessFailures of autonomously operating AI agents in settings where humans cannot intervene before harm occurs.
- Strengthen AI incident monitoring and emergency protocolsOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Scale up public funding for AI safety and governance researchGovernance & Oversight Untracked
Recommended by: Governing AI: A Plan for Canada , Governing AI: A Plan for Canada (2024 Update)
- Establish joint international limits on AI in critical domainsGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Monitor AI adoption risks in critical domainsOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Encourage reliable AI diffusion and adoptionOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
General & machinery mitigations
Mitigations not tied to a single tracked risk — government machinery, broad capacity building, and preparedness.
- Adopt antitrust measures against ASI monopolyGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Establish a central federal AI governance bodyGovernance & Oversight Untracked
Recommended by: Governing AI: A Plan for Canada , Governing AI: A Plan for Canada (2024 Update)
- Enact comprehensive risk-based federal AI legislationGovernance & Oversight Untracked
Recommended by: Governing AI: A Plan for Canada , Governing AI: A Plan for Canada (2024 Update) , Liberal Party 2026 Convention resolution: a Canadian Artificial Intelligence Act
- related: Bill C-27 — Digital Charter Implementation Act, 2022 (including AIDA) — AIDA (part of Bill C-27) attempted a risk-based framework and died on the Order Paper when Parliament was prorogued in January 2025.
- Coordinate ASI use across countries to avoid conflictGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Build trust to limit military AI race escalationGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Ensure inclusive international ASI governanceGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Strengthen information-sharing protocolsOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Develop systems to address widening international income inequalityGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Champion and fund global AI governance effortsGovernance & Oversight Untracked
Recommended by: Governing AI: A Plan for Canada , Governing AI: A Plan for Canada (2024 Update) , Preparing for the AI Crisis: A Plan for Canada
- Limit AI in military decisions; regulate autonomous weaponsGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Launch a national public conversation on AIGovernance & Oversight Untracked
Recommended by: Governing AI: A Plan for Canada , Governing AI: A Plan for Canada (2024 Update) , Preparing for the AI Crisis: A Plan for Canada
- Develop protocols for predictable AI-enabled military escalationGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Develop redundant knowledge repositoriesOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Remove regulatory impediments to AI adoptionGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Prepare mechanisms for safe international ASI cooperationGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Build shared understanding of global AI risksGovernance & Oversight Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
- Invest in skills, data, and computeOperational Process Untracked
Recommended by: AI and National Security: Scenarios Workshop Summary Report
Not yet mapped
Recommendations from tracked reports that have not been linked to a canonical mitigation yet.
- Pivot to meet the AI crisis