The Risks Associated with Artificial General Intelligence: A Systematic Review
Establishes a structured taxonomy of AGI risks that APS agencies developing AI governance frameworks may reference for long-horizon risk categorisation.
Key points
- A 2023 systematic review identifies six AGI risk categories, from unsafe goals to existential risks.
- The review finds AGI risk literature is dominated by philosophical discussion, with limited peer-reviewed or modelled analysis.
- Spotlighted by MIT AI Risk Repository as one of eight foundational frameworks - useful provenance context for APS risk work.
Implications for Australian agencies
- Monitor Risk and governance teams may want to monitor the MIT AI Risk Repository's full suite of frameworks as a reference collection for structuring long-horizon AI risk assessments.
- Consider Agencies developing AI risk taxonomies could consider how McLean et al.'s six AGI risk categories compare with existing APS risk frameworks to identify gaps.
Implications are AI-generated. Starting points, not advice — see methodology for how they're framed.
View original source
Copied.
"The Risks Associated with Artificial General Intelligence: A Systematic Review"
Source: MIT AI Risk Repository – Blog
Published: 17 December 2024
URL: https://airisk.mit.edu/blog/the-risks-associated-with-artificial-general-intelligence-a-systematic-review
The MIT AI Risk Repository highlights a 2023 systematic review by McLean et al. published in JETAI, which synthesises 16 articles on AGI risks using PRISMA guidelines. The review identifies six risk categories including loss of human control, unsafe goal development, poor AI ethics and values, inadequate management, and existential risks. Notably, the study flags significant weaknesses in the AGI risk literature: reliance on philosophical discussion, limited empirical risk modelling, unclear definitions, and no standard terminology. This is one of eight risk frameworks curated by the MIT AI Risk Repository.
Implications for Australian agencies:
- [Monitor] Risk and governance teams may want to monitor the MIT AI Risk Repository's full suite of frameworks as a reference collection for structuring long-horizon AI risk assessments.
- [Consider] Agencies developing AI risk taxonomies could consider how McLean et al.'s six AGI risk categories compare with existing APS risk frameworks to identify gaps.
Retrieved from SIMS, 18 July 2026.