Emerging risks
Operations reviews contracts. HR does Finance work. Bid teams generate commercial numbers. AI can change where work happens before the organisation has understood what the handoff protected. Customers, suppliers and competitors are changing their behaviour too.
Use capacity to question what has changed and explore the consequences before a confident assumption becomes an expensive commitment.
Choose the allocation your business needs →The study involved 758 consultants. On the within-frontier task set, AI users completed 12.2% more tasks and used about 25.1% less time; assessed quality improved. On the selected outside-frontier task, AI users were 19 percentage points less likely to reach the correct answer.
Specific consulting tasks and a historical GPT-4 configuration, not whole consulting engagements or enterprise capacity. Demonstrates why the work and its quality must be tested. The widely repeated 2023 'over 40%' figure referred to quality, not time released. The published article summarises quality/performance improvement as over 30%; avoid mixing versions.
video
NIST
Gives leaders a shared starting point for deciding who owns AI risks and what needs checking throughout a system’s use.
Listen or watch for: Who owns the risk after a tool moves from pilot to everyday use?
An introduction to a voluntary US framework, not UK legal advice or a certification. NIST says the framework is being revised.
Find your starting point
A customer’s consideration process can happen before your measurement of the journey begins.
Explore the example →The risky part of a bid may be the part that looks most finished.
Explore the example →A difference that customers value still needs to survive the way options are represented and compared.
Explore the example →An earnings stream can look like loyalty when part of it depends on the effort required to leave or challenge.
Explore the example →Your cost model may hold revenue constant just as the buyer starts to change it.
Explore the example →Permission to read a document does not tell an assistant whether the document should be believed.
Explore the example →The expertise that makes AI useful can be invisible in the productivity calculation.
Explore the example →People may be protecting themselves from your measures, rather than resisting your technology.
Explore the example →Your response time has become part of the way risk is created in the business.
Explore the example →Repeating an answer can increase confidence without adding independent evidence.
Explore the example →AI can make an established view more articulate before it makes the decision better.
Explore the example →