PromptRiskDBThreat intelligence atlas
AI Risk

Nascent capabilities (agency and autonomy)

Example: "Connection to a code interpreter or email server can result in unintentional harm if, while trying to fulfill a request by the user, a model performs tasks beyond what the user has asked for. For example, a user seeking a job may ask a model to provide detailed information on a potential employer. A model with adequate connectivity and excessive agency may attempt to fulfill that request by not only gath...

View related techniquesRead profile
AI Risk

Record summary

A quick snapshot of what this page covers.

Techniques0Attack methods connected to this risk.
Mitigations0Defenses that may help with related attacks.
Domainn/aThe broad risk area this belongs to.

Risk profile

How this risk is described and categorized.

Example: "Connection to a code interpreter or email server can result in unintentional harm if, while trying to fulfill a request by the user, a model performs tasks beyond what the user has asked for. For example, a user seeking a job may ask a model to provide detailed information on a potential employer. A model with adequate connectivity and excessive agency may attempt to fulfill that request by not only gathering information from the web but also emailing current employees or the CEO of the company to request they answer questions."

Domainn/a
Subdomainn/a
Entityn/a
Intentn/a
Timingn/a
CategoryEthical and social risks
SubcategoryNascent capabilities (agency and autonomy)

Related techniques

Attack methods connected to this risk.

No linked attack methods. No AI attack method is connected to this risk in the current data.

Suggested mitigations

Defenses that may help with related attacks.

No propagated mitigations. No defense is available through the connected attack methods.

Source

Research source for this risk, when available.