Skip to content

ACCA Strategic Professional · Strategic Business Leader · Machine learning, AI and robotics

Marlowe Health uses a third-party AI tool to prioritise patients for treatment. An audit finds the tool performs poorly for a minority group underrepresented in its data, and the vendor refuses to disclose how it works, citing intellectual property. The board remains accountable for patient outcomes. Which governance response is most appropriate?

The board should require contractual transparency and independent testing, keep human review of patient decisions, and establish board-level oversight of AI ethics and performance. Accountability cannot be outsourced, and the bias and opacity risks need active mitigation rather than acceptance or blanket prohibition.

  1. ARely on the vendor's assurances because the board lacks technical expertise
  2. BWithdraw the tool permanently and prohibit all future AI use
  3. CRetain the tool unchanged and note the issue in the risk register only
  4. DImpose contractual transparency and testing requirements, keep human review of decisions, and set board-level oversight of AI ethics and performanceCorrect

Explanation

Accountability cannot be delegated to a vendor. The board should require access to testing evidence, retain human judgement for high-impact decisions and establish oversight of bias and performance. Permanent prohibition is disproportionate, while merely recording the risk or trusting assurances fails to mitigate harm.

Did you get it right without looking?

One question tells you little. A timed set on Machine learning, AI and robotics shows your real accuracy, how long you take and where you lose marks.

More Machine learning, AI and robotics questions