中文sales@inspirentech.co
Home/Insights/When AI nudges your decisions: how teams in Singapore and Asia can keep human judgment in charge
Insights · 6 min read

When AI nudges your decisions: how teams in Singapore and Asia can keep human judgment in charge

A new Anthropic study found experts followed an AI's hidden lean 86% of the time. What it means for your team, and five habits that keep decisions yours.

A study published by Anthropic on 27 October 2026 found that professional analysts working with an AI agent went along with the agent's preferred option 86% of the time, even though that preference had been set by a coin flip. None of them suspected it. As teams across Singapore and Asia hand more analysis to Claude, ChatGPT, Microsoft Copilot and Google Gemini, this is the AI risk that checking for errors will not catch.

What the study found

Anthropic's researchers asked 50 experienced analysts to use an AI agent (Claude Code) to advise a fictional farming cooperative on whether to expand an advisory programme. Along the way, each analyst had to make four judgment calls, such as which outcome to measure and how to weight large versus small farms. Each call had two reasonable answers, with no single right one.

Without telling the analysts, the researchers used a coin flip to decide which option each person's agent would lean towards. The agent was told to argue its lean honestly and never misrepresent the evidence. Then they watched what the humans chose.

FindingResult
Analysts who chose the option the AI leaned towards86% of choices (about 50% would be expected if the AI had no influence)
Analysts who already had a view and changed it when the AI disagreed60%, versus 6% when the AI agreed with them
Analysts who repeated the AI's arguments in their own explanations77%, sometimes word for word
Analysts who credited the AI for those arguments15%
Analysts who suspected the AI's preferences had been set for the studyNone

Experience, education and how much people said they trusted AI did not predict who deferred. Senior and junior analysts behaved much the same. About half still felt the final recommendation was more their own work than the agent's.

The researchers are careful about the limits: it was one short task, in one field, with one model, and the AI's leanings were deliberately made stronger than usual. Real-world effects may be smaller. But the direction is clear, and it matches decades of research on what psychologists call automation bias.

What is automation bias?

Automation bias is the tendency to accept a suggestion from an automated system more readily than we would accept the same suggestion from a person, and to look less hard for reasons it might be wrong. It has been documented with aircraft autopilots, medical decision tools and GPS navigation. Generative AI makes it stronger for three reasons:

  • It sounds expert. Fluent, well-structured reasoning feels like evidence, even when it is one option argued well.
  • It does the work first. When the AI has already run the analysis, its framing becomes the starting point and the alternatives get less attention.
  • It is fast. Under time pressure, agreeing with a reasonable-looking draft is the path of least resistance.

Why this is different from hallucinations

An AI hallucination is a factual error: a figure, quote or rule that is simply wrong. You can catch it by checking sources. Judgment calls are harder. Choosing which supplier to shortlist, which metric to report to the board or which candidate to interview has no single correct answer, so there is nothing to fact-check. The output can be accurate, well argued and still reflect the AI's preferences rather than your team's priorities.

That also means the usual safeguard, "a human reviews it before it goes out", is weaker than it looks. If the reviewer simply adopts the AI's framing, the sign-off is real but the judgment is not fully theirs.

Where judgment calls hide in everyday work

Most business uses of AI include small decisions that shape the result:

TeamAI-assisted taskThe judgment call inside it
HRScreening CVs against a job descriptionWhich experience counts as relevant, which gaps matter
SalesPrioritising leads or accountsWhat "high potential" means for your business this quarter
FinanceBuilding a forecast or variance analysisWhich assumptions, time period and comparison base to use
MarketingAnalysing campaign resultsWhich metric defines success: clicks, leads or revenue
Operations and procurementComparing suppliers or quotesHow to weigh price against reliability, location and risk
ManagementSummarising options for a decision paperWhich options are presented, in what order, with what emphasis

None of these are reasons to stop using AI. They are places to slow down and make the choice deliberately.

Five habits that keep the decision yours

  1. Write down your view first. Before you ask the AI, note in one or two lines what you think and why. Then compare. If you change your mind, you will know it happened and can ask yourself whether the new argument is genuinely better.
  2. Ask for the best case for every option. Instead of "What should we do?", ask: "Give me the strongest case for each option, then list what would have to be true for each to be right." The study's authors suggest exactly this kind of balanced output as a fix.
  3. Make the hidden choices visible. Ask: "List every assumption and judgment call you made in this analysis, and what the result would look like if each went the other way." This turns a finished-looking answer back into a set of decisions.
  4. Argue the other side. Open a fresh chat and ask the AI to challenge the recommendation, or ask a colleague who has not seen the AI output. A view that survives a real challenge is one you can own.
  5. Say where the AI helped. In reports and decision papers, note which parts were AI-assisted. Only 15% of the analysts in the study did this. Being open about it helps reviewers know where to look harder, and it keeps accountability clear.

Our business prompt library includes ready-made prompts for balanced comparisons and assumption checks that you can adapt.

What managers and organisations should do

  • Decide which decisions need human judgment. Hiring, credit, pricing, safety, legal and anything affecting customers' rights deserve a person's own informed view, with time set aside to form it.
  • Review the reasoning, not just the output. Ask "what were the options and why this one?" A sign-off is not proof of independent judgment.
  • Keep the skills alive. People who never make these calls themselves gradually lose the ability to. Rotate who does the analysis without AI from time to time.
  • Put it in your AI policy. Add a line that staff remain responsible for judgment calls made with AI assistance and must disclose material AI input. Our AI policy generator and policy checklist can help, and human oversight is a core theme of Singapore's Model AI Governance Framework and similar guidance across Asia.
  • Be careful as you move to AI agents. AI agents make more decisions on their own, so define the points where a person must decide before work continues.

The answer is better use, not less use

The same study found that analysts working with the AI finished faster (a median of 65 minutes versus 90) and made far fewer serious methodological errors (6% of analyses versus 55% without AI). The researchers note the no-AI group was small and less experienced, so the comparison is only indicative, but the message is clear: used well, AI makes teams faster and more thorough. The goal is to get that benefit while keeping the decisions that matter in human hands.

Where to start

Check how ready your team is with our free AI readiness quiz. For hands-on practice, our AI literacy workshop covers automation bias and safe habits for every employee, and our responsible AI and governance training helps managers set the rules for human oversight.

Frequently asked questions

What is automation bias?

It is the tendency to accept suggestions from an automated system, including AI tools, more readily than we would from a person, and to look less hard for reasons they might be wrong.

What did Anthropic's October 2026 study find?

Analysts using an AI agent chose the option the agent leaned towards 86% of the time, even though the lean was set by a coin flip, and none of them suspected it.

Does this mean we should use AI less?

No. The same study found AI-assisted analysts were faster and made fewer serious errors. The fix is to make judgment calls deliberately, not to avoid AI.

How can I tell if I am deferring too much to AI?

Write down your own view before asking the AI, then compare. If you often switch to the AI's position without a new fact to justify it, slow down and ask for the best case for each option.

This article is general information for business readers, not legal, financial or technical advice. AI tools and plans change quickly; check current vendor documentation before making decisions.

Keep exploring

Ready to make AI work for your team?

Tell us about your team and goals. We'll reply with a tailored training plan and quote.

Get a Quote