All issues

Intelligence Brief

Week of
November 9, 2026

Your weekly edge, the research, frameworks, and signal that separate operators from spectators. Read and summarized so you do not have to.

Published Monday, November 9 at 6:00 AM PT. 12 sources, 9 minute read.

01

The Human Intelligence Edge

What makes humans irreplaceable in the AI era.

Nature Human BehaviourOctober 26, 2026

Generative AI and the temporary upskilling of knowledge workers

Emma Wiles et al.

Why it matters. 487 consultants, randomized, with AI lifting performance on data-science tasks outside their existing skill set and no independent skill remaining once the tool was removed. This ran in last week's collaboration section and it belongs here too, because the cleanest part of the result is not about the pairing at all. It is a measurement of what the person kept. One profession and one study window, so the size will move, but the shape is the thing to plan against: assisted output and acquired capability are indistinguishable in a finished deliverable and behave completely differently the moment the tool is unavailable. Any claim about a team getting stronger has to be tested with the assistance switched off, or it is a claim about the tool.

Deloitte and Oxford EconomicsMarch 4, 2026

2026 Global Human Capital Trends

Deloitte Insights with Oxford Economics

Why it matters. More than 9,000 leaders across 89 countries, with adaptability, creativity and judgment named as the differentiators, and organizations taking human-centric approaches to AI more likely to beat their own return expectations. It is a leader survey, so it reports what executives believe about their workforce rather than what their workforce can do. The useful part is where it puts the line. Human capability is being described as a performance lever with a number attached to it, not as culture work that happens beside the P&L. That reframing is only worth something if the hiring process can actually see adaptability and judgment, which a resume and two interviews cannot. Decide what evidence counts before the next requisition opens.

McKinsey Global InstituteJanuary 6, 2026

Human skills will matter more than ever in the age of AI

Alexis Krivkovich and Anu Madgavkar

Why it matters. Thousands of workplace skills mapped across job postings, with more than 70% relevant to both automatable and non-automatable work and 12% still human-only. The 12% is the number people quote and the 70% is the one that matters. Almost every skill you hire for now sits on both sides of the line, which means the question stopped being whether a skill survives automation and became how a person combines it with the tool. Postings describe what employers asked for, not what the work required, so treat the proportions as a map rather than a forecast. For a hiring process the instruction is concrete: stop sorting skills into safe and doomed piles, and start testing how a candidate applies judgment, critical thinking and relationships to work the model has already touched.

02

The Broken System

Why resumes, interviews, and hiring infrastructure are failing.

Harvard Business ReviewNovember 2, 2026

Research: When Workers Adapt Their Resumes to Fit the Political Moment

Jay Fitzgerald, on research by Joseph Pacelli, Janet Gao, and Jun Oh

Why it matters. 2.9 million US workers analyzed, with resume language shifting to match the prevailing climate and that signaling influencing hiring outcomes even where predicted job fit was slightly lower. The study reads observed language against observed hiring, so the causal arrow is inferred rather than assigned. What it documents is a market that pays for phrasing. Candidates learned which words land this year and rewrote themselves accordingly, which is rational behavior and ruinous as a selection signal, because the skill it rewards is reading the room on a page nobody will ever check. The counter is not better resume screening. It is a step in the process where the candidate does the work and somebody who watched them do it goes on record.

ACL EMNLP 2026October 2026

Bias in the Tails: How Name-conditioned Evaluative Framing in Resume Summaries Destabilizes LLM-based Hiring

Huy Nghiem, Phuong-Anh Nguyen-Le, Sy-Tuyen Ho, and Hal Daume III

Why it matters. Close to a million generated resume summaries, where the factual content stayed largely stable and the evaluative wording moved with race and gender name cues. Generated summaries in a research setup are not your ATS, and the effects sit in the tails rather than the average. The finding still lands on the exact thing most hiring audits are built to check. A fairness review that confirms the facts were carried over correctly will pass a summary that calls one candidate promising and another adequate on identical evidence. Framing is what a human reader acts on. If a model touches candidate text anywhere in your funnel, the audit has to read the adjectives, not just reconcile the dates.

arXiv preprintOctober 30, 2026

AI evaluators favor job applications written by certain AI models, independent of candidate quality

Nicolo Pagan, Michael Andreas Ebling, Stefania Ionescu, and Aniko Hannak

Why it matters. 5,500 AI-written cover letters and 414,000 model evaluations, with scores shifting according to which model drafted the application while the underlying CVs were held constant. Simulated pipeline, preprint, so hold the magnitudes loosely. The mechanism is the part worth acting on, and it echoes the self-preferencing result from two issues ago. A candidate's score is picking up their tool choice, which is a function of budget, habit and whatever their last employer paid for, and neither the applicant nor the recruiter can see it in the output. Everything written before the interview is now a measurement of the writing stack. Weight the part of your process the model cannot submit on the candidate's behalf.

03

The SupaHuman Standard

The behavioral science of who actually thrives.

Behavioral SciencesNovember 4, 2026

From Psychological Resilience to Employment Quality: A Four-Wave Longitudinal Study of Career Adaptability and Job-Search Self-Efficacy

Yang Zhang and Hewen Wu

Why it matters. 860 employed graduates tracked across four waves, with resilience positively associated with employment quality and the proposed routes through career adaptability and job-search self-efficacy testing weaker than the direct link. Four waves is a real design and the sample is early-career graduates in one country, so the ceiling on generalization is low. Note which part held. The straightforward association survived and the mechanism story did not, which is the opposite of how this literature usually reports. For assessment that is a simplification rather than a loss: you do not need the causal pathway to decide that how someone handled a genuinely bad stretch is worth asking about in detail, with dates and names and what they tried second.

Frontiers in PsychologyNovember 3, 2026

Emotional intelligence and teaching practicum performance among Chinese preservice physical education teachers: The mediating role of teaching self-efficacy

Xin Wang and Zhihuan Zhang

Why it matters. 462 student teachers, with self-reported emotional intelligence associated with mentor-rated practicum performance and teaching self-efficacy accounting for part of that link. Cross-sectional and confined to one discipline in one country, so nobody should port the coefficients anywhere. The design detail is what earns it a slot: the outcome came from a mentor who watched the person work, not from the same questionnaire that measured the trait. That is rarer than it should be in this literature, and it is the structure to copy. Self-rated interpersonal skill tells you how a candidate sees themselves. Someone who supervised them through a hard week tells you what actually happened, and those two readings diverge often.

BMC PsychologyOctober 23, 2026

A three-wave serial mediation model: linking emotional intelligence to subjective career success via career adaptability and proactive behaviors

Haci Arif Doganulku and Hazel Duru

Why it matters. 315 working professionals across three waves, where emotional intelligence predicted later career adaptability, which linked to proactive behavior and then to subjective career success. Self-reported throughout and the final outcome is satisfaction rather than anything an employer counts, so read it as a sequence rather than a result. The sequence is the useful part. Adaptability showed up before the proactive behavior did, which says the thing to look for is not a general willingness to act but evidence that someone recalibrated first and moved second. In an interview those look identical when described and completely different when walked through: ask what changed their read on the situation, and when.

04

The Future of Work, Without the Clichés

Where AI and human collaboration is actually heading.

Frontiers in PsychologyOctober 16, 2026

A study of the impact of AI-employee collaboration on employee innovation behavior

Yan Guanyu and Niu Chenchen

Why it matters. 259 employees across three waves, with AI-employee collaboration predicting innovation behavior partly through self-efficacy, and servant leadership strengthening the path. One organizational context, self-reported measures, so the effect size is not portable. The conditional is, and it keeps appearing: the gain ran through how people felt about their own capability under the tool, and it got larger where the manager's job was defined as clearing the way. Neither of those is a procurement decision. A team can have identical model access and identical prompts and land on opposite sides of this result depending on whether somebody is accountable for making the use of it work. Tool rollout is not the variable being measured here.

PNASSeptember 29, 2026

Human learning is an understudied but promising lever for boosting human-AI synergy

Julian Berger et al.

Why it matters. A reanalysis of 74 studies, where human and AI teams tended to beat the better solo partner when people received both feedback and AI explanations, and explanations without feedback were associated with negative synergy. It is a meta-regression over other people's designs, so it describes a pattern rather than establishing cause. Sit with the negative half. Explaining the model's reasoning to someone who never learns whether they were right is the default configuration of most AI tooling shipped this year, and it tracked with teams performing worse than the stronger side alone. The missing piece is cheap and nobody builds it: tell people what the outcome was. A workflow with no feedback loop is not a collaboration, it is a handoff.

World Economic Forum with AccentureMarch 2026

Organizational Transformation in the Age of AI: How Organizations Maximize AI's Potential

Fatima Gonzalez Novo, Jill Hoang, and Karen O'Regan

Why it matters. Cross-industry cases and consultation with a community of more than 450 executives, turning collaboration from a slogan into three design decisions: who is accountable by name, how much autonomy the system gets before a human is required, and what the escalation path is when it exceeds that. A practice synthesis built on self-selected cases, not a controlled study, and the examples skew large-enterprise. The decisions still transfer down to a team of eight. Most organizations have never written any of the three down, which is why nobody can say who owns an AI-assisted output when it turns out wrong. Answer the three in a paragraph each, and most of the governance argument stops being abstract.