This hub collects articles on AI law, ethics, safety, and open questions about the future of AI, as they apply in Canada and the United States. It is written for readers who want sourced summaries of what rules, courts, and researchers have actually said.
Key takeaways
- AI rules differ by country, and some frameworks are binding while others are voluntary.
- Copyright, patent, and deepfake questions have different legal status in the US and Canada as of October 2026.
- Risk claims range from documented current harms to long-term forecasts, and the articles separate the two.
- No AI is accepted as conscious, and no agreed test for AI sentience exists.
- Rules and official guidance
- Copyright, patents, and contested uses
- Fraud, surveillance, and security
- Ethics frameworks and trustworthy AI
- Education, work, and transport rules
- Risk, safety, and the threat debate
- Consciousness, sentience, and emotion
- Reasoning, capability, and the future of AI
Rules and official guidance
Binding AI rules differ by jurisdiction, and several are still changing. The European Commission describes the EU AI Act as sorting AI into four risk levels with a list of prohibited practices (European Commission). The articles below cover Canada, the EU, US federal practice, courts, and cyber agencies. This page is general information, not legal advice.
- Canada AI Regulation: Where Federal and Provincial Law Stand
- EU AI Act Risk Tiers, Bans, Fines, and the 2026 Delay
- Unethical or Useful? What Federal Rules Ask of High-Impact AI
- Why AI Emotion Recognition Gets People Wrong
- AI Hallucinations in Court Filings: Case Counts and Rules
- Securing AI Systems: What Cyber Agencies Advise in 2026
- What Counts as a Dominant AI Company Under Competition Law
Copyright, patents, and contested uses
Ownership of AI-assisted work and the use of training data are active legal questions in the US and Canada. These articles set out what agencies and courts have said, with the status noted as of October 2026. Statements of law are summaries of public sources and not legal advice.
- AI and Copyright: Who Owns AI-Generated Content in 2026?
- What the U.S. Copyright Office Says About AI Training
- AI-Generated Art Ethics: Training, Credit, and Court Cases
- Who Owns an AI-Assisted Invention? US and Canadian Patent Rules
- Three AI Policy Disputes: Copyright, Deepfakes, Weapons
- AI Sexual Deepfakes: US TAKE IT DOWN Act and Canada's C-16
Fraud, surveillance, and security
Documented misuse of AI includes cloned voices, facial recognition errors, and automated attacks. These articles rely on regulator, agency, and court records rather than anecdotes, and also cover what governments have pledged on nuclear weapons.
- AI Voice Cloning Scams: What Records Show and Which Laws Apply
- Which AI Is Most Dangerous? Facial Recognition Cases Checked
- AI and Cyber Security: Who Gains, Attackers or Defenders?
- Why AI Threat Detection Struggles With False Alarms
- AI and Nuclear Weapons: What Governments Have Committed To
- AI in Cybercrime: What the FBI's 2025 Complaint Data Show
- Paid AI Stock Lists: What SEC and CSA Notices Require
- Can AI Forecast Crypto Prices? What Two 2026 Studies Found
Ethics frameworks and trustworthy AI
AI principles often conflict with one another, and some ethics documents are voluntary while others are binding. The NIST AI Risk Management Framework, for example, is described by NIST as intended for voluntary use (NIST). The articles compare frameworks from NIST, the EU, Canada, UNESCO, WHO, and religious bodies.
- AI Ethical Dilemmas: Where Principles Conflict and Rules Apply
- The Ethical Dilemma of AI: Can Machines Be Given Morals?
- What Trustworthy AI Means in NIST, EU, Canadian, and ISO Texts
- Is AI Inherently Unethical? A Documented Case Tested
- What Human-Centered AI Means: Shneiderman, NIST, Stanford
- Explainable AI Classes: Self-Interpretable, Local, and Global
- UNESCO's Generative AI Guidance: A Checklist for Ministries
- What the WHO Says About AI Ethics in Health
- AI and Christianity: What Evangelical and Anglican Bodies Say
Education, work, and transport rules
Sector rules decide how AI is handled in schools, workplaces, and shipping. These articles cover detector errors, school trust, education regulation, workplace safety evidence, and rules for autonomous vessels.
- AI Detectors in Schools: False Positives and Fair Process
- Trust Between Students and Teachers When AI Is in the Room
- How Education Is Regulated: ESSA in the US, Provinces in Canada
- AI and Workplace Safety: What the Evidence Shows So Far
- Crew Rules and Accident Data for Autonomous Ships
- Autonomous Ships and Maritime Law: IMO and Canadian Rules
- Transport Canada's Rules for Small Autonomous Vessels
- How US Insurance Regulators Oversee AI Use by Insurers
- AI and Cannabis in Canada: What the Act, Audit, and Data Show
Risk, safety, and the threat debate
Claims about AI risk range from documented current harms to long-term loss-of-control arguments. These articles separate the two, using the 2026 international safety report, incident databases, forecasting results, and public opinion surveys.
- Is AI a Threat to Humanity? The Evidence on Both Sides
- Is AI a Threat or a Tool? What the 2026 Evidence Shows
- AI vs Humanity: What a Forecasting Tournament Found
- How Researchers Sort AI Risks: The MIT Risk Repository
- Is AI a Curse? What Incident Records Can Show
- Is AI Broken? How AI Systems Fail and How to Check Them
- Can an AI Release Be Undone? Open Weights Explained
- Elon Musk on AI Risk and Regulation: The Documented Record
- AI Pros and Cons: Where Experts and the Public Disagree
- Is AI a Blessing? What Canadian and US Polls Show
- AI Domination: Who Holds Control, Based on 2026 Data
Consciousness, sentience, and emotion
No AI is accepted as conscious, and no test for AI sentience exists. These articles explain what researchers measure, which theories disagree, and where expert opinion remains open.
- When AI Becomes Aware: How Researchers Would Tell
- When Will AI Become Sentient? What Surveys Actually Show
- Can AI Become Self-Aware? What the Research Shows
- Can AI Experience Emotions? Detection, Imitation, and Feeling
- Does Artificial Intelligence Dream? What the Science Says
- AI Sentience vs Animal Sentience: What Evidence Counts
- Could AI Be Conscious? Where Two Leading Theories Disagree
Reasoning, capability, and the future of AI
Whether AI reasons, improves itself, or surpasses people depends on definitions and on the task measured. These articles look at benchmark data, evolutionary methods, and the origins of the field.
- Can AI Think for Itself? Reasoning, Autonomy, and Oversight
- AI Reasoning vs Human Reasoning: What Tests Show
- Can AI Reason? What Math Tests With Changed Details Show
- Can AI Build and Improve Itself? What the Evidence Shows
- Can AI Evolve? Evolutionary Algorithms, Tested
- What Counts as AI Surpassing Human Intelligence?
- Can AI Exceed Human Ability? What the 2026 AI Index Measured
- Can AI Outsmart Humans? What Test Scores Do and Do Not Show
- Why AI Was Created: What Turing and Dartmouth Proposed
- Is There a Best AI Robot? Shipment Data and Test Results
- How AI Invents Objects It Never Saw in a Picture
FAQ
How does the EU AI Act classify AI systems?
The European Commission describes four levels: unacceptable risk, high risk, transparency risk, and minimal or no risk. Prohibited practices include emotion recognition in workplaces and education institutions.
Is the NIST AI Risk Management Framework mandatory?
No. NIST describes the framework as intended for voluntary use. Version 1.0 was released on January 26, 2023.
What rules apply to automated decisions by the Canadian federal government?
The Treasury Board Directive on Automated Decision-Making applies to federal departments that use automated systems to make or support administrative decisions. It requires an Algorithmic Impact Assessment and a way for people to challenge decisions.
Has the U.S. Copyright Office addressed AI?
Yes. Part 2 of its AI report, on the copyrightability of outputs, was published on January 29, 2025. Part 3, on generative AI training, was released in pre-publication form on May 9, 2025.
Is any AI conscious?
No AI is accepted as conscious, and researchers disagree about how such a claim could be tested. The articles on consciousness and sentience cover the main theories.
Related reading
More in AI Blog.
Last verified: October 2026