Dual Use Science risks

Capabilities and Risks from Frontier AI

DSIT (2023)

Source

Sub-category

Risk Domain

4Malicious Actors & Misuse

4.2Cyberattacks, weapon development or use, and mass harm

Using AI systems to develop cyber weapons (e.g., by coding cheaper, more effective malware), develop new or enhance existing weapons (e.g., Lethal Autonomous Weapons or chemical, biological, radiological, nuclear, and high-yield explosives), or use weapons to cause mass harm.

"Frontier AI systems have the potential to accelerate advances in the life sciences, from training new scientists to enabling faster scientific workflows. While these capabilities will have tremendous beneficial applications, there is a risk that they can be used for malicious purposes, such as for the development of biological or chemical weapons."(p. 22)

Entity— Who or what caused the harm

Human

Due to a decision or action made by humans

AI system

Due to a decision or action made by an AI system

Other

Due to some other reason or is ambiguous

Intent— Whether the harm was intentional or accidental

Intentional

Due to an expected outcome from pursuing a goal

Unintentional

Due to an unexpected outcome from pursuing a goal

Other

Without clearly specifying the intentionality

Timing— Whether the risk is pre- or post-deployment

Pre-deployment

Occurring before the AI is deployed

Post-deployment

Occurring after the AI model has been trained and deployed

Other

Without a clearly specified time of occurrence

Supporting Evidence (3)

"Frontier AI models can provide user-tailored scientific knowledge and instructions for laboratory work which can potentially be exploited for malicious purposes.194 Studies have shown that systems may provide instruction on how to acquire biological and chemical materials.195"(p. 23)

"While our focus is on frontier AI, it is important to note that frontier capabilities can be used in conjunction with the capabilities of narrower biological AI design tools,200 such as AlphaFold2201 and RFDiffusion.202 Narrower AI tools can already generate novel proteins with single simple functions and support the engineering of biological agents with combinations of desired properties.203"(p. 23)

"Existing frontier AI models have some ability to design and troubleshoot laboratory experiments.196 These capabilities can be enhanced when equipping LLMs with the ability to access tools such as web search and specialised computational tools.197 When connected to laboratory robots or cloud labs,198 LLMs can directly instruct these platforms to carry out experiments.199"(p. 23)

Part of Loss of control