brianletort.ai
← Library

Boston Consulting Group

Verified evidenceCreator to judge

GPT-4 integrated into individual consulting task flows

That access to a general model improves all apparently similar knowledge tasks.

Engineering design · Global BCG consultant population

Collections: Negative results

An editorial scene for Boston Consulting Group contrasts analyze source materials, generate ideas, and draft recommendations unaided by gpt-4. with delegate or interleave ideation, analysis, and drafting with the model. in the gpt-4 integrated into individual consulting task flows workflow.

Executive brief

The operating-model shift, in one view.

The operating control is task classification, not generic prompt training; users need a way to recognize when the model is outside its frontier.

AI value · Task completion speed and correctness by task type

Verified

Inside the frontier: 25.1% faster, 12.2% more tasks, over 40% higher quality; outside the frontier: 19 percentage points less likely to be correct

Experimental tasks approximated consulting work but were not client delivery; the negative out-of-frontier result is as important as the positive averages.

Before

Analyze source materials, generate ideas, and draft recommendations unaided by GPT-4. → Check coherence and submit the work.

After

Delegate or interleave ideation, analysis, and drafting with the model. → Verify, revise, and submit model-assisted work.

Human boundary

The model proposes content and analysis; the consultant owns verification and the submitted answer.

Why it matters

That access to a general model improves all apparently similar knowledge tasks.

How the work changed

Before

How the work ran before the change.

  1. Step 1 of 2

    Consultant

    Analyze source materials, generate ideas, and draft recommendations unaided by GPT-4.

  2. Step 2 of 2

    Consultant

    Check coherence and submit the work.

What changed

That access to a general model improves all apparently similar knowledge tasks.

Decision rightHuman moves from creator to judge

After

How the same work runs now.

  1. Step 1 of 2

    Consultant with GPT-4

    Delegate or interleave ideation, analysis, and drafting with the model.

    ControlUser prompting and task choice

  2. Step 2 of 2

    Consultant

    Verify, revise, and submit model-assisted work.

    ControlHuman final decision

Process model built from the published workflow evidence for Boston Consulting Group. Every step, actor, and control appears in full below.
Every step, actor, and control

Exception path

For tasks outside the model's capability frontier, consultants must solve independently or validate against source evidence rather than accept the model's confident answer.

Work removed

  • Some first-pass ideation, analysis, and drafting on tasks inside the model's capability frontier

Decision authority

The model proposes content and analysis; the consultant owns verification and the submitted answer.

Before

  1. 01

    Consultant

    Analyze source materials, generate ideas, and draft recommendations unaided by GPT-4.

    Control: Professional judgment

  2. 02

    Consultant

    Check coherence and submit the work.

    Control: Human accountability

After

  1. 01

    Consultant with GPT-4

    Delegate or interleave ideation, analysis, and drafting with the model.

    Control: User prompting and task choice

  2. 02

    Consultant

    Verify, revise, and submit model-assisted work.

    Control: Human final decision

Work that left the path

  • Some first-pass ideation, analysis, and drafting on tasks inside the model's capability frontier

Human role before

Consultants performed every analytical and drafting step directly.

Human role after

Consultants decide when to delegate, when to integrate, and when to distrust the model.

AI roleGeneral-purpose GPT-4 assistance across realistic consulting tasks.

Outcomes

Task completion speed and correctness by task type

Verified

Randomized no-AI control consultantsInside the frontier: 25.1% faster, 12.2% more tasks, over 40% higher quality; outside the frontier: 19 percentage points less likely to be correct

2023 preregistered experiment · 758 BCG consultants, about 7% of individual-contributor consultants

Experimental tasks approximated consulting work but were not client delivery; the negative out-of-frontier result is as important as the positive averages.

What leaders can reuse

Anti-pattern

Applying the positive average to all knowledge work or assuming prompt training eliminates frontier risk.

Questions

  1. 01How will users identify out-of-frontier tasks?
  2. 02What evidence must be checked before submission?

Portability conditions

  • Observable task types
  • Source-based verification
  • Human ability to withhold delegation

Reputation risk

low

Evidence and authority

What the public record supports.

Current · updated

2 independent; publication outcomes are verified.

Bundle 1.0.0 · reviewed 2026-08-23 · stable ID 0dc0a03188456b2e

Related transformations

More in Engineering design

Sources

Read the evidence, freshness, caveat, and version policy.