Harvard Business Review just launched a new reading list on hiring and developing talent in the age of AI, and three of the editor-recommended articles land on the same uncomfortable truth:
AI hasn't removed the hard parts of people decisions, it's relocated them. Candidates can now perform competence they don't have. Managers are losing the coordination work that used to define their jobs. And "fairer" algorithms are quietly redefining what fairness even means. Below is a closer look at each article, and where PI Hire and Obi already address the gap.
1. AI Has Broken Hiring. Here's How to Fix It.
The premise here is blunt: generative AI has broken the proxies hiring has relied on for decades. Researchers spent five months interviewing 120 talent-acquisition leaders and analyzed 6,380 recorded screening calls. Their finding was that screens now reward the performance of competence more than competence itself. A polished résumé, a smooth recorded video answer, a take-home assessment quietly run through an AI model — none of that measures capability anymore. It measures how well someone can produce a good artifact, with or without help.
That's a structural problem, not a candidate-honesty problem. If the test is a static document or a scripted, unmonitored response, AI fluency will always outperform authentic ability at getting a candidate through the door. The article's fix is to move early-stage hiring toward authentic reasoning, judgment, and adaptability — live problem-solving, real-time reasoning, and interpersonal signals that are much harder to fake or outsource to a model in the moment.
Where this connects to PI Hire: this is precisely the failure mode that validated, static-trait assessment was built to avoid. The PI Behavioral Assessment and PI Cognitive Assessment don't ask candidates to perform a task that can be polished offline with AI assistance — they measure stable behavioral drives and cognitive ability directly, under conditions designed to resist coaching and gaming. Job Targets built from top performers give hiring managers a validated benchmark instead of a gut read on how "polished" someone seemed. In an environment where interview performance is increasingly manufactured, a validated, harder-to-game data point matters more, not less.
2. How AI Is Redefining Managerial Roles
This piece draws on a Harvard Business School study led by Manuel Hoffmann, tracking over 50,000 global software developers from 2022 to 2024, half of whom used GitHub Copilot. Researchers logged 2.4 million actions and split them into core work (coding) and managerial work (coordination, status updates, oversight). The pattern: coding output rose about 5%, while project-management activity fell about 10%. Gen AI is absorbing the coordination tasks that used to justify a layer of middle management, flattening hierarchies and pushing employees toward more autonomous work.
The catch the article raises is that this shift doesn't happen neutrally. Someone has to decide what gets automated and who captures the upside — more capacity for higher-value work, or just fewer managers and thinner support for the people still doing the work. Left undirected, AI adoption concentrates benefit rather than distributing it.
Where this connects to Obi and PI's management tools: if AI is stripping out the administrative half of a manager's job, what's left is the human half — coaching, feedback, conflict navigation, and knowing how to reach each person on their team. That's exactly the gap Obi by PI is built for. It doesn't replace managerial judgment; it prepares it, generating a tailored playbook — grounded in a specific employee's behavioral data — before a tough feedback conversation, a first 1:1, or a conflict between two people. As gen AI absorbs coordination work and shrinks the traditional scope of the middle-manager role, the managers who keep their edge will be the ones equipped to do the remaining, harder job well: leading actual people. That's the case for pairing AI-driven efficiency gains with tools like Obi rather than assuming managers will just figure out the people side on their own.
3. New Research on AI and Fairness in Hiring
This one complicates the usual debate. One camp argues AI reduces human bias and noise in hiring decisions; the other argues it can amplify existing inequities at scale. A three-year ethnographic study of a global consumer-goods firm, by van den Broek, Sergeeva, and Huysman, found both camps are missing something: adopting an algorithmic hiring system doesn't just apply a definition of fairness, it locks one in. The firm's system privileged a rigid, one-size-fits-all notion of fairness that pushed out hiring managers' local judgment about context, role, and team fit. The result was narrower candidate pools and frustrated managers who no longer felt they had a say.
The researchers' core point is that fairness isn't a property baked into code — it's continually negotiated by the people who design, deploy, and use the system. Their advice to leaders: ask which definition of fairness your system is actually encoding, who has the authority to set or challenge it, and whose judgment gets quietly sidelined as the system scales.
Where this connects to PI Hire: this is the strongest argument for a validated, transparent methodology over a black-box algorithm. PI's assessments are built on decades of validation research with published, auditable criteria — not a model that infers hidden patterns from historical hiring data (and risks re-encoding whatever bias lived in that data). Job Targets are set deliberately, by the people who know the role, using defined behavioral and cognitive criteria — not derived silently by an algorithm optimizing for a definition of "fit" no one explicitly chose. That keeps the hiring manager's judgment in the loop by design, which is exactly the ingredient the HBR study found missing.
The Common Thread
All three articles point at the same underlying shift: AI is removing the friction from tasks that used to force careful human judgment — writing a strong interview answer, coordinating a project, defining what "the right candidate" or "a fair process" means. Removing friction is good. Removing judgment is not. The organizations that come out ahead won't be the ones with the most AI in their hiring and management stack; they'll be the ones that used AI to remove the busywork while deliberately protecting the validated, human-grounded judgment underneath it — which is the exact model PI Hire and Obi are built around.
If you want to talk through how these shifts apply to your hiring pipeline or management practices specifically, we're happy to walk through it.
