Talent intelligence is the current rave. Several vendors are working on or already offering talent planning data. Their offer is seductive. They promise that if you feed AI the org chart, financial info, a skills inventory, if you have one, and some market data, you will get a workforce plan. That plan will include a recommendation for how many people you need, with what skills, in which locations, at what cost, eighteen months from now.

The vendors offering this break into three camps. Workday leads on connected, scenario-driven financial planning; Oracle and SAP offer suite-native depth; Visier, Eightfold, Gloat, and Reejig cover skills intelligence, analytics, and mobility. Add Draup and the former Skyhive for job-to-automation estimation. But how good are the results? And can Ai really do this now?

Not close, yet, and maybe never.

The First Problem is the Data
Workforce planning runs on data. Bad data produces bad plans, and most companies have bad data. I have not encountered one firm with good, verifiable data about its workforce or its skills.

Start with the skills inventory. Almost no organization knows what skills its people have. The HRIS has job titles and pay grades, but it does not record the quiet data. It doesn’t know that the financial analyst taught herself Python last year or that the warehouse supervisor ran a logistics operation in the military. Skills data is self-reported, out of date, or absent. The taxonomies are not consistent. One system calls it “project management,” another “team leadership,” and a third does not capture it at all.

AI does not fix this. AI just inherits whatever data there is, whether good or bad. And, a model trained on garbage produces polished garbage.

The talent intelligence everyone wants requires a clean, current, structured view of the workforce that very few companies possess. Until that exists, AI makes assumptions, and you cannot tell which data or numbers are real and which are invented.

External labor market data is better, but still messy and incomplete. Job posting aggregators, compensation benchmarks, and skills demand signals exist and improve every year. But they describe the past and the present. They do not know your business is about to enter a new market or shut a product line. The most important inputs to a workforce plan are decisions that have not yet been made.

The Second is That Planning is Not Prediction
Workforce planning follows strategy, and strategy is based on human judgment about the uncertain future.

How many engineers you need next year depends on what you decide to make. What you decide to make depends on where you think the market is going, what your competitors do, what capital you can raise, and what your leadership has the appetite to attempt. These are not data problems. Or AI problems. They are judgment problems, full of politics, ego, and incomplete information.

The most significant limitation of AI is its complete blindness to non-codified knowledge. Strategic corporate planning relies heavily on the unwritten: the boardroom negotiations, impending executive transitions, politics, cultural appetites for risk, or confidential M&A considerations, and what leadership thinks the market will do.

An executive leadership team may intentionally hire too many in a specific region to signal market commitment to investors or to deliberately deprive a competitor of talent. They may decide to cut positions to force efficiency or cut costs. These decisions are not recorded in a database. They exist solely in human conversations and strategic intent. An AI model, blind to these unquantifiable variables, will recommend unachievable plans based on this perfectly imperfect data.

An AI model can take “we will grow the European business forty percent” and tell you what that means in terms of headcount needs and cost. That is useful and achievable. What it cannot do is decide whether to grow the European business forty percent or whether that is possible or wise. Scenario generation is only as good as the scenarios you feed it. Again, garbage in, polished garbage out.

The job of AI is not to generate multiple futures and recommend an optimal one. The future scenarios come from the business.

What AI Can Do
None of this means AI is useless. It means the useful applications are narrower and less exciting than the promise.

Read more

Most of the conversation about AI runs on industrial age assumptions. We talk about jobs. Productivity. Labor. Efficiency. Augmentation. All of it is surface stuff and first-order effects. The deeper question is what is left that is distinctly human when intelligence and knowledge stop being scarce?

We have always defined ourselves partly through consciousness and language. We were the planners. The pattern spotters. The strategists. The ones who reasoned in symbols. That was why we thought we were special.

AI is rewriting that story. Not perfectly, not quite yet, but the direction is set, and it does not reverse.

We assume human value is whatever AI cannot do yet. But that notion is a slippery slope. First goes Chess. Then translation. Coding. Research. Medical diagnosis. Every one of those was once proof of human uniqueness. One by one, they fell.

So the future is not about protecting human exclusivity. It is about rewriting what we mean by participation, meaning, agency, and consciousness itself. That is a far bigger thing than a shaken-up job market.

The deep questions can wait. The next ten years are going to hurt long before anyone gets philosophical about it.

This wave is different from the ones we keep comparing it to. Factory automation and farm mechanization came for physical work first and gave people decades to climb into offices. This time it lands on the office. Drafting, summarizing, first pass analysis, code, research, document review, and routine customer contact. That is the work that pays the mortgages of the educated middle class today, and it is exactly the work that gets cheap and automated first.

Nobody fires the whole department on a Monday. That is not how it goes. As we can already see, companies stop backfilling roles that are left. They slow down graduate hiring. They quietly expect the team of eight to do what the team of twelve used to do. In some cases, it is mass layoffs of lower-level workers; in other cases, it shows up as a hiring freeze rather than a layoff, which is why it is easy to miss until you are the one not getting hired.

The first rung that breaks is the bottom one. Entry-level white-collar work is what AI now does at a fraction of the cost. That creates a problem nobody has solved. If the machine does the junior work, where do the seniors come from? You cannot develop experienced judgment without the years of grunt work that built it. We are about to find out what happens to a profession that eats its own pipeline.

For anyone in talent acquisition, the change is direct. The job stops being about filling requisitions faster and starts being about deciding what the workforce should even look like. Fewer reqs. More questions about which work needs a human at all, which gets handed to a system, and how you keep an organization learning when the apprenticeship layer is thinning out.

Call it the efficiency purge. Not a collapse, but a steady reweighting of who and what an organization is willing to pay for. It runs unevenly, faster in some sectors and places than others, across five to fifteen years rather than five to fifteen months. The mistake is treating it as a temporary disruption you ride out. It is a structural reset.

That is the near term. It is real, it is already moving, and it is still only the first-order effect.

AI is not a tool
We still talk as if there are two separate players at work. Human plus machine. Worker plus tool. But a powerful enough AI does not act like a tool in any sense we know. There may not be a partnership between human and machine. Or whatever partnership there is may be temporary.

AI will likely become a layer, an extension of memory and perception. A system that evolves alongside us and shapes how we think. That changes humans as much as it changes work.

Go PAID and get it ALL

We can already see the rough early versions of this change. People offload memory to their devices. They outsource judgment to recommendation engines. They let AI write and think for them. They hand emotional contact to algorithms. They build part of their identity inside digital systems.

Future AI looks less like software and more like the infrastructure for thinking. Infrastructure is the stuff you build on top of without noticing. You do not “use” it as a separate act. It just underwrites everything else you do. When it works, it is invisible. When it breaks, everything sitting on it stops.

In the future, you would no more think that someone “used AI to write an essay” than you would think someone “used the alphabet to write an essay.” AI is a substrate, not an app.

What survives
So, what part of being human survives when intelligence is everywhere and free?

Start with the body. We live inside physical reality through sensation, mortality, pain, aging, desire, and plain biological limits. AI can model all of that. It does not live it.

Then there is consciousness. We still cannot explain why subjective experience exists at all. Intelligence and consciousness may be two different things. AI could blow past us intellectually and still have nothing going on inside. Or it could grow something conscious that is nothing like ours.

It may also become or act as an alien. People keep imagining AI growing up to be like us. More likely, it develops nonhuman forms of reasoning and perception. No emotion. Massively parallel. Running on a different clock. Not located anywhere in particular. Indifferent to survival or status. At home in concept spaces that we cannot picture. That makes it less like a person and more like a separate category of mind.

Most people still think about AI in terms of its effects on the economy and work. The bigger story is that it is closer to a new civilization.

If intelligence becomes infinite, cheap, everywhere, and always on, then every institution built on scarce expertise wobbles. The industrial world was organized around scarce labor. The knowledge economy is organized around scarce expertise. AI removes or reduces the value of both. It removes the need for management, professional status, credentials, corporate hierarchy, government administration, and even the whole idea of a career identity.

In every past technological revolution, we survived by climbing. Physical labor to cognitive labor. But AI climbs the entire cognitive ladder with us, and then past us. The human job shifts toward interpretation, ethical judgment, experience, relationships, and deciding what the goals should even be. Less doing the task. More deciding which tasks are worth doing and why.

The hard part is psychological. We get our identity from being useful. If AI eventually takes away most of the things the market pays for, the crisis is not economic. It is existential. What is a human life for if not productivity?

That is the question that will define the rest of this century. If survival no longer depends on labor, we either rediscover what industrial life trivialized or we drift into passivity and dependence on the algorithm.

The deciding factor is probably not how capable AI gets. It is whether we keep our hands on the wheel while the systems around it take shape. These shifts move unevenly and across generations. We are at the start of a long transition, not anywhere near the end.

So the important question was never “which jobs survive AI?” It is this. What stays uniquely meaningful about being human once we share the world with a mind that is not human at all?

Share

Leave a comment

Go PAID and get it ALL

I take on a few advisory clients a year. Mostly, firms want someone to tell them the truth about the talent market rather than what they hope to hear. If that is useful to you, email me DM me..

For as long as I can remember, there has been a gentle tug of war between the hiring manager and the recruiter. From a recruiter's perspective, the hiring manager has been on the edge of the process. They owned the requisition, had the budget, and made the decision about who to hire. But the work of hiring, sourcing, screening, scheduling, and negotiating was the recruiter's responsibility. The hiring manager critiqued. The recruiter delivered or didn’t.

That arrangement is ending.

AI agents can now do most of what recruiters used to do. And over the next few years, they will take over virtually all search, screening, assessment, scheduling, at least early-stage negotiation, and offer preparation. They will do it faster, more cheaply, and more precisely than any human. By 2030, the recruiter, as a facilitator/coordinator, will largely be gone. What will remain is a small senior layer of strategists, advisors, and governance specialists. The operational work will belong to the agent.

But this article is not about recruiters losing their jobs. It is about the hiring manager finally being forced to do what should have always been their job.

The End of the Alibi
The hiring manager’s traditional attitude was based on a useful alibi. If a hire went wrong, the recruiter sent the wrong people. If the role stayed open too long, the recruiter was slow. If diversity numbers slipped, the recruiter had not sourced widely enough. The hiring manager had clean hands and little responsibility.

But with the recruiter gone, the alibi disappears. The AI agent does not get blamed in the same way. The agent does what it is told to do based on the inputs it receives from the manager, and a handful of senior advisors who can offer guidance when needed. When the agent underperforms, the responsibility belongs to the hiring manager.

The hiring manager of 2030 cannot stand on the edge of the process. They are the process.

What Does the New Hiring Manager Actually Do?

Read more

This is a fictional story about the looming future of work, HR, and recruiting. I used the childhood characters Dick, Jane, Spot, and Puff to link the recent past to the near future. It shows how age…

Read more

When we talk about AI and recruiting, we discuss automation, efficiency, bias reduction, sourcing tools, assessment algorithms, and productivity gains. We ask whether AI can screen resumes more accur…

Read more

Lots of TA people and hiring managers are thinking but not saying out loud that skills-based hiring is the most overhyped idea in TA right now. It has been for three years. And the only reason it has remained popular is that it seems to be a differentiator from the usual credentials and experience that most jobs require. And many very credible voices are too invested in it to admit what the data shows.

Harvard Business School and the Burning Glass Institute tracked actual hiring patterns and found that fewer than one in 700 hires in 2023 benefited from the shift to skills-based hiring. It isn’t any better today. Meanwhile, 85% of companies claim they practice it. This is a movement built almost entirely on press releases and lots of conference babble.

The Policy May Have Changed. The Behavior Did Not.
Quite some time ago, IBM dropped degree requirements. Then Accenture. Walmart rewrote its job descriptions. Bank of America announced changes. Sixteen states have removed four-year degree mandates for most public sector roles. Maryland was first in 2022. Pennsylvania, Virginia, and New Jersey followed. Tennessee wrote it into law.

All of that is good. Removing credential bias from job postings opens the door to many more candidates and improves diversity. A degree requirement on a job posting that does not need one automatically screens out 76% of Blacks and 83% of Latinos before anyone even looks at the resume.

But the problem is that the postings changed, but the hiring managers did not. Nor did many recruiters and HR folk.

When a job is posted without a degree requirement, the same hiring manager still reviews the resume. They still have the same mental model of what a qualified candidate looks like. They still ask the same interview questions. They still make the same offer to the person who looks most like the last person who did the job well. The policy changed. The behavior did not. And no one is measuring the behavior or working to change it.

Colorado is one of the few exceptions worth studying. The state government tracked actual outcomes and found that 25% of hires in 2023 and 2024 within reclassified job categories were people without degrees. That is what implementation looks like. But it required executive orders, hiring manager training, and a workforce board running workshops with employers on how to write different job descriptions. Most companies announce a policy change and call it done with no training or enforcement.

The Vendor Promise Is Also Broken
The pitch from companies like Eightfold, Gloat, and Beamery is that you can build a skills-based operating model for your workforce. Map skills to roles. Identify gaps. Move internal talent to where the work is. Hire based on verified capability rather than credentials. It sounds great; how can anyone object?

The problem is that this approach assumes you have clean, structured, and verified skills data for your workforce and candidates. Almost no organization has that. What vendors are doing is pattern-matching job titles to inferred skills and calling it talent intelligence. An engineer has “Python” on their profile because the job description said Python, not because anyone verified that they can actually write it. The candidate has “project management” listed because everyone does. There is little to no verification, assessment, or testing

The problem is the gap between what is promised and what is delivered.

Two Problems Dressed Up as One
The deeper issue is that skills-based hiring mixes two entirely different problems.

Problem one: reducing credential bias in hiring. This can only work when degree requirements are removed from job postings, hiring managers are trained to evaluate differently, and there is a way to track actual hiring outcomes. This is time-consuming work and requires the ability to influence and to enforce compliance.

Problem two: building a skills-based operating model for talent. This means developing the whole architecture of how you classify work, define roles, move people, and measure capability based on actual skills. This is a transformation of how organizations think about labor. It requires verified skills data, new job architectures, manager behavior change at scale, and compensation systems built around capability rather than title. It is a different and much harder problem.

The hype has merged these two problems so thoroughly that companies feel like they are addressing the second by doing the first. They drop the degree requirement. They feel progressive. They do not track or seriously evaluate whether any of it changed who actually got hired.

The Amplification Machine
Part of why this has lasted for three years without any serious pushback is the institutional weight behind it. The World Economic Forum has been pushing skills-based hiring as a core pillar of the future of work agenda. LinkedIn’s Economic Graph team constantly publishes data on skill trends. The Tear the Paper Ceiling coalition, backed by Opportunity@Work, runs campaigns and gets major corporate signatories. McKinsey has written about it. These are credible, well-resourced organizations. What CHRO or CEO is going to challenge these sources?

When so many credible voices agree, it usually means the idea is correct in principle but much harder in practice than anyone is saying. That is what is happening. Credential bias is real. Degree inflation is real. The talent pool exclusion is real.

But the idea, while solid in principle, is failing by its own metrics. The University of Phoenix says that for every 100 job listings that remove degree requirements, fewer than four candidates without degrees are actually hired.

What Actually Works

Read more

Last week, I presented an argument a CEO might make for eliminating or greatly reducing the internal TA function. This week, how a TA leader or CHRO might respond.

Speaking to the CEO:

You are right t…

Read more

What is the probability of a CEO forcing the reduction or outsourcing of the current talent function to qualified and vetted RPOs, specialized external recruiters, and AI Agents?

I place this probability at 80-85% over the next 18 months for the following reasons. In many firms, this transition is already underway.

The Internal Resistance Is Real but Overrated
Every TA leader will make the case for keeping the function internal. They’ll talk about culture, institutional knowledge, employer brand, and the value of human judgment in hiring. Some of that is genuine, but most of it is self-preservation presented in what they think is strategic language. A CEO who has watched AI grow and become embedded in finance, legal, and administrative work is not going to find that argument as compelling today as it was in 2019.

The TA function has always struggled to prove its strategic value in terms that demonstrate effectiveness, not just activity. As I have written before, time-to-fill, cost-per-hire, and offer acceptance rates are operational metrics, not business impact metrics. If the CHRO cannot walk into the boardroom and demonstrate in dollar terms what the internal TA function produces that an RPO cannot, the function is already losing the argument.

The CEO Case for RPO Is Structurally Strong Right Now
Several conditions converge to make this a serious option, not just a cost-saving one.

Variable cost in an uncertain environment
With hiring mostly stagnant and the long-term outlook murky, a large fixed-cost TA team is a liability. RPO converts the fixed cost to variable. You pay for what you use. In a world where the CFO is scrutinizing every headcount line and AI is promising efficiency everywhere else, the TA function’s cost structure is hard to defend. A well-structured RPO contract tied to hiring volume is a good model for uncertainty.

The RPO market has matured
This isn’t 2005. Qualified enterprise RPO providers such as Cielo, Korn Ferry RPO, Allegis, and AMS all operate at Fortune 100 scale with sophisticated technology, data capabilities, and compliance infrastructure. They can almost always match or beat your argument that “they can’t match our quality. Many RPOs have built and are using AI agents for sourcing, screening, and more. By leveraging AI Agents and technology, they reduce costs, improve speed, and increase both their service and candidate quality.

Specialized recruiters solve the high-judgment problem
The only valid argument for keeping recruiting internal applies to senior, specialized, or culture-critical hires. But that alone does not justify the cost. It still makes sense to outsource 80% of volume hiring to an RPO while maintaining a small internal team that uses AI Agents and develops executive search relationships for the top of the house. This is not a compromise but a rational division of labor.

AI makes the handoff cleaner
The more of the sourcing and screening workflow that AI handles, the less advantage internal teams have. If an RPO can plug into your ATS, run AI agents, and operate within your compliance framework, the integration costs that historically made RPO transitions painful are declining.

Where the CEO Hesitates
But I am not predicting that this scenario is a 100% probability for a few reasons.

Control and data
Talent data and competitive intelligence are strategic assets. It is critical to know who you’re talking to, what the competitive talent landscape looks like, and who you have in your internal pipeline. Providing that knowledge to a third party makes the firm dependent on that party to maintain privacy and data security. It requires a high level of trust. A sophisticated CEO will want a strong contract giving clarity on data ownership before signing.

Culture at scale is real
Employer brand is vital to attract the right talent. RPO partners can be briefed and managed, but the candidate experience is harder to control at arm’s length. If external suppliers fail in a high-visibility hiring situation, it can cost more in employer brand damage than the operating costs saved.

This article continues below. Next week, I will present this case from the perspective of a CHRO. Be sure to subscribe to have access to the CHRO’s views on the probability of outsourcing TA.

Subscribe now

Read more

Many TA leaders measure their importance and power by the size of their team. More headcount means more budget, more influence, more organizational power. A thirty-person recruiting department needs a director, but a five-person team only needs a manager. In a traditional corporation, this makes a difference.

But in the emerging new world, a large team is a death sentence.

The large, in-house recruiting department is dying. And the leaders who create a new model, one borrowed from operations, are the ones who thrive.

The model that replaced manufacturing should replace recruiting
In the 1970s, most large manufacturers owned the entire production process. They made the parts, assembled the product, and ran the warehouses. Vertical integration was normal. Then lean operations theory, followed by global supply chains, turned that logic inside out. By the 1990s, the most efficient manufacturers had learned to own the design and the relationships, not the manufacturing or the labor. Semiconductors are designed in the US and built in Taiwan; Nike does not make shoes. Apple does not build iPhones.

But the TA and HR functions are still operating under the manufacturing model, where they own everything. They staff for peak demand and measure activity as if that showed their effectiveness.

The result is a function that has too many recruiters when hiring is slow and not enough when hiring demand peaks.

What the new model might look like
The future is about developing and nurturing a small, senior team. I call them Talent Architects. Their job is not to recruit. Their job is to design, source, orchestrate, and measure a recruiting supply chain that does the recruiting for them.

That supply chain has three main components.

The first is specialized external recruiters engaged on an as-needed basis. A network of specialists, technical sourcers, executive search practitioners, and diversity-focused boutiques, each used for specific roles and time-defined engagements. You pay for output. You pay for fills. You do not pay for availability.

The second is RPO capacity that scales with volume. The right RPO relationship is not a blanket outsourcing arrangement. It is a defined engagement covering specific job families, geographies, or hiring cycles. When volume drops, the engagement scales down. There are no conversations about carrying costs.

The third, and the one most TA leaders have not yet taken seriously, are AI agents. Not AI-assisted recruiters. AI agents operating as first-line sourcers, screeners, and scheduling coordinators. More top-performing teams are leveraging these agents to handle high-volume screening workflows without human intervention. The next generation of agents will take over passive sourcing and initial outreach. They will be able to screen candidates and provide them with feedback. These agents are not tools for recruiters. They are recruiters. They just do not have paychecks.

The Talent Architect sits above all three. They design the intake process. They set the quality bar. They manage the vendor relationships and the AI configurations. They own the data. They are accountable for outcomes.

The metrics have to change
A supply chain model requires supply chain metrics. And supply chain metrics have a clear boundary. Recruiting owns the process, but the business owns the employment relationship. The moment a hire walks in the door, the recruiting supply chain’s job is done.

This is not a dodge. Holding recruiting accountable for retention, performance, or engagement is like holding a manufacturer responsible for how the buyer uses the product. The variables after handoff are no longer yours.

Read more

Most recruiters are still trying to convince themselves that the automation and AI wave won’t affect them. After all, they are the people people, and they will always have a job. But you’ve already done the math. Now what?

You know too much to feel comfortable, and not enough people above you know you know it.

Your current role feels safe. But a company that has frozen or eliminated internal growth and is quietly reducing headcount is not a safe harbor. The comfort you feel is the comfort of the known, not the comfort of the secure.

You’ve watched your ATS evolve from a filing cabinet into something that screens, scores, schedules, and sometimes interviews candidates without a human-in-the-loop. You’ve seen your company’s sourcing volume handled by tools that don’t sleep, don’t ask for headcount, and don’t need a LinkedIn Recruiter license. You’ve read the same data that everyone else in your field is ignoring: employers cut planned new hires by 35% in 2025. The companies still hiring are doing it with fewer recruiters, not more.

But here’s what separates you from most of your peers: you understand what actually survives the automation wave. Not resume parsing. Not pipeline management. Not scheduling, screening, or status updates because those are already gone or in progress.

What survives is the work that requires genuine human judgment, trust, and relationships. Advising hiring managers who don’t know what they really need. Building talent communities before a requisition opens. Managing the ecosystem of agencies, universities, and professional networks that no algorithm can fully own. Being the person a candidate trusts with a career decision, not just a job opening.

You can do that work. You’re already thinking at that level.

You are likely sitting at your desk right now, looking at a calendar of interviews or a pipeline of candidates, feeling a profound sense of "stuckness." You are excellent at what you do. You have the experience, the intuition, and the results to back up a promotion. Yet, you’ve hit a ceiling that feels less like glass and more like reinforced concrete. Above you, your managers are parked. They aren’t moving up because there is nowhere to go, and they aren’t moving out because they are just as terrified as you are.

So let’s talk about what you actually do with that.

A paid subscription to this Substack is not simply content access but a gateway to four decades of insight and experience across recruiting, human resources, and organizational development. It consolidates frameworks, trend analysis, and practical guidance into a single, continuously evolving resource designed for practitioners operating in a rapidly shifting talent landscape.

Subscribe now to gain immediate access and stay ahead of the structural changes reshaping how organizations attract, assess, and deploy talent.

Subscribe now

Read more

Last week, Meta laid off another round of employees. Second time this year. They hit Reality Labs, Facebook operations, sales, and recruiting. Recruiting was hit hardest.

Read that again. The company, with $200 billion in annual revenue and growing cloud revenue while posting record profits, cut its recruiting team first.

Not legal. Not finance. Not the layers of middle management that every CEO publicly claims to despise.

Recruiting.

The Pattern Is Hiding in Plain Sight
Here is how AI restructuring works at big companies.

First, the company spends 12 to 18 months investing heavily in AI infrastructure.

Second, leadership conducts an internal audit of which roles can be automated or reduced. They look at functions that are cost centers with no discernible strategic value.

Third, the layoffs arrive, blaming AI and framed as a competitive necessity.

Fourth, recruiting is among the first teams to be reduced because leadership has already concluded that hiring is slowing and that AI tools can absorb the rest.

From an executive perspective, if you are hiring fewer people, you need fewer recruiters. If AI is screening applications, you need fewer coordinators. If robotic agents are sourcing and doing outreach, you need fewer sourcers.

We have seen 45,000 to 60,000 tech jobs eliminated in Q1 2026 alone. That is a 51% increase over Q1 2025. Of those cuts, roughly 20% are being explicitly attributed to AI by the companies themselves. The other 80%? Cost restructuring, post-pandemic correction, and Wall Street pressure for margin improvement.

But recruiting gets cut in either case, because recruiting is seen as a cost function tied to hiring volume. When volume drops, the function does too.

The Uncomfortable Question Nobody Is Asking
We have spent years arguing for a seat at the table. We have done everything we can think of, from strategic workforce planning to developing quality-of-hire metrics to establishing a talent intelligence function. We talk about how indispensable we are, but executives don’t buy it. Recruiters are seen as transactional, easily replaced when needed.

Companies cut the recruiting team before the analytics team, before the marketing team, and in some cases before the teams that haven’t shipped anything in two years.

So, we have tried to build the case for strategic value, but executives still see it as a cost center.

While some TA leaders have genuinely transformed their functions into workforce intelligence operations that leadership cannot imagine operating without. Most have not.

Most are still measured on time to fill and requisition load, which are metrics that describe activity and throughput, not impact.

When the AI restructuring wave arrives, executives cut throughput functions first. They protect judgment functions.

What Is Actually Happening to the Talent Pipeline
There is a secondary crisis developing underneath the headline layoff numbers that recruiting leaders should be loudly raising right now. Entry-level job postings on platforms like Handshake are down 15% year over year. Applications per vacancy are up 30%. Companies are eliminating the roles that feed leadership pipelines, and nobody in the C-suite is modeling what the organization will look like in five years when there is no internal bench.

This is TA’s argument to make. Not “we need more budget for sourcing tools.” The argument is: the people being cut today are the senior leaders of 2030. Eliminating entry-level roles in the name of AI efficiency is placing a bet that AI will produce senior judgment when you need it. There is no evidence that is true.

73% of TA leaders in Korn Ferry’s 2026 data say critical thinking is the skill they need most. AI skills rank fifth. The judgment required to lead organizations does not come from a prompt. It comes from years of experiencing failure and getting feedback inside an organization. You cannot source it later. You have to grow it now.

What TA Leaders Should Do Differently
I have argued for years that our normal metrics are rubbish. Stop measuring yourself by hiring volume and activity metrics. Volume and speed metrics make your function look like a cost center.

Read more

As I said in last week’s article, we are standing in the wreckage of the recruiting process. Artificial intelligence has increased the efficiency, scale, and speed of hiring processes, yet it has also led to the loss of meaningful human connection and increased candidate dissatisfaction.

The question is not whether to use AI in recruiting. That question has been answered.

The more important question is how to design a recruiting model in which AI handles scale and process optimization, while human interaction is available and used where it creates the greatest value.

This is the foundation of a high-touch, AI-enabled candidate experience.

The Core Problem: Efficiency Without Connection
No global recruiting function can operate efficiently without A)-based systems to increase speed and throughput. But these systems cannot make the process human. They cannot identify or reveal a candidate’s judgment, intent, adaptability, or cultural alignment.

As I said last week, at its core, recruiting is an information problem, but not all information can be quantified or measured. are often revealed only through human interaction. When we use AI, these interactions are minimized or eliminated, and recruiters and hiring managers risk making decisions based on incomplete or distorted data.

A New Design Principle: Human Interaction as a Strategic Resource
The solution is not to go back to fully manual processes. That would be both impractical and inefficient.

Instead, organizations must treat human interaction as a scarce and valuable resource, deploying it intentionally where it really counts in the candidate journey.

This requires a shift from a process-centric model to an experience architecture model. Rather than asking, “Where can we automate?” the more relevant question becomes, “Where does human interaction create the greatest value?”

There are three critical moments or times in a candidate’s journey that are critical:

1. Moments of uncertainty – where candidates need clarification, reassurance, or context

2. Moments of evaluation – where nuanced judgment is required from both sides

3. Moments of decision – where commitment, negotiation, and alignment occur

We need to design a high-touch experience at each of these points.

Practical Strategy 1: Humanize the First Meaningful Interaction
AI is highly effective at initial sourcing and outreach, but the transition from automated engagement to human contact is often poorly designed. Many candidates interact with bots or automated emails for extended periods before speaking with a recruiter.

This is a missed opportunity.

The first meaningful human interaction should occur early enough to establish credibility and trust, but not so early that it disrupts scale. A practical approach is to:

• Use AI for initial identification and qualification

• Trigger a human interaction once a candidate meets predefined relevance thresholds

• Ensure that the first human conversation is meaningful, not transactional

This interaction should focus on understanding the candidate’s motivations, constraints, and career context, rather than simply validating resume data. It sets the tone for the entire experience.

Practical Strategy 2: Redesign Screening as a Hybrid Process

Read more

We are standing in the wreckage of the recruiting process while trying to pretend that what is happening is just an evolution. But AI, along with other factors, has wrought chaos.

What we are seeing is a classic technological arms race.

Candidates use AI to optimize applications. Employers deploy AI to detect or filter them. Candidates then refine their tools to evade those filters. Each cycle increases the game's sophistication while reducing the reliability of the signals produced.

Arms races rarely produce clarity. They produce escalation.

Here is where we are in 2026. Candidates are submitting AI-optimized resumes tailored to job descriptions in minutes. Cover letters are generated in seconds. Video interviews are being coached in real time by AI tools whispering answers through an earpiece. Work samples are assembled with AI assistance. And on the employer side, ATS systems are auto-rejecting candidates before a human ever looks at them, AI tools are scoring video interviews based on tone and word choice, and hiring managers are ghosting finalists who made it to the last round.

Hiring has always been an information problem.

Employers are trying to infer future performance from incomplete and imperfect signals: resumes, references, interviews, and credentials.

What AI has done is dramatically increase the noise in the system. Candidates can now generate signals that appear to demonstrate capability but are artifacts of software assistance. At the same time, employers have automated screening systems that amplify superficial indicators such as keywords or linguistic patterns.

The result is predictable: signal collapses, noise expands, and trust erodes

Both sides are gaming the other. Both sides know it. And the entire process has become a ritual that nobody believes in anymore.

So what does a recruiter do?

The Signal Problem
The honest answer is that most of what you receive in the top of the funnel is now close to useless as a measure of candidate quality. This is not the candidate’s fault. When you auto-reject someone for using the word “managed” instead of “led,” you trained them to stop being honest and start being strategic. You built the conditions for this. The resume they send you now is not a document about them. It is a document about what your ATS wants to see. Those are very different things.

The numbers bear this out. In the past three years, total applications per role on LinkedIn have doubled. At the same time, 66% of recruiters say finding quality candidates has gotten harder. More volume, less signal. That is what an arms race looks like when it matures.

The video interview is not much better. A candidate coaching tool called Final Round AI, which provides real-time AI assistance during interviews, reportedly had 300,000 users within months of launching. When nearly a third of hiring managers say they have encountered fake voices or AI-scripted answers in interviews, you must ask what you are evaluating. You are no longer assessing the candidate. You are assessing how well they have prepared their AI to impersonate them.

Emerging Adaptations
Smart recruiters are shrinking the funnel, not expanding it.

Instead of processing

Read more

Note: This is a longer article than usual because it discusses where work is heading and how what has happened in the AI job marketplace foretells what is coming for all of us.

For more than a century, the job description has been the foundational unit of work organization. It was a model built on the assumption that work is stable enough to be described in advance. That work consisted of a set of duties, reporting relationships, qualifications, and expected outputs that remained consistent over time, and that this set of duties equaled what we call a “job.” That assumption is crumbling.

And if you want to see where it crumbles fastest, watch what is happening to the AI labor market right now.

Consider the title “AI Engineer.” It did not exist as a recognized role category five years ago. Today, it commands a median total compensation exceeding $245,000 in the United States, with top earners at frontier companies pulling $875,000 or more.

Netflix posted a single AI role with a salary ceiling of $900,000. Anthropic and OpenAI have structured packages for senior researchers that reach eight figures over multi-year terms. These are not anomalies in an otherwise stable compensation landscape. They are the leading edge of a structural rupture.

What makes the AI labor market instructive is not the numbers themselves but the logic behind them. Companies are not paying $800,000 for a job. They are paying for a specific, rare, time-sensitive capability to build and deploy systems that will reshape their entire cost structure. The capability is everything. And that same logic, once loosed from the frontier of AI, will migrate steadily into the rest of the economy.

A Tale of Two Labor Markets
The AI employment market has already split the world of work into two sharply diverging tracks. In 2025, tech companies laid off nearly 131,000 workers while simultaneously struggling to fill AI and machine learning positions. Microsoft eliminated thousands of jobs and raised AI engineer salaries past $300,000. Amazon cut positions across multiple business units and, within weeks, posted hundreds of openings for data scientists and ML specialists.

Human resources departments whose entire function is organized around job descriptions saw job postings decline by 16% as the sector was itself restructured. AI-related HR postings, by contrast, surged 25% in the first quarter of 2025 alone, with median salaries hitting $157,000.

This is not a contradiction. It is a preview. Companies are eliminating roles defined by stable, specifiable task sets that can be written into a job description, while hiring capabilities that are fluid, rare, and compounding in value. The job description, as a technology, was built for the first category of work. It has almost nothing useful to say about the second.

Younger workers are learning this lesson the hard way. Entry-level positions have declined 29% since January 2024. The unemployment rate for college graduates aged 22–27 reached 4.8% in mid-2025, meaningfully above the overall rate, as the traditional entry-level-to-executive career ladder built entirely on the logic of job-based progression fractured under the weight of AI-driven task automation. A degree in hand and a job description to match no longer adds up to a career path.

From Roles to Tasks, Skills, and Projects

Read more

When I first wrote about humanoid robots and recruiting in mid-2025, the honest answer was that the technology was impressive but mostly just staged. The deployments were only pilots, heavily supervi…

Read more

In 2022, we had the “Great Resignation.” In 2024, we had “Quiet Quitting.” In February 2026, we have entered what I’m calling the “Great Compliance.” According to recent survey data, only 7% of workers say they would quit over a return-to-office mandate today, down from 51% just a year ago.

But this isn’t because employees have suddenly rediscovered their love for fluorescent lighting and cubicles. It’s because they’re terrified of what we discussed last week: the hiring recession that has left job seekers competing for what few jobs are still available while corporate America rakes in millions in profits.

The leverage has shifted. It is no longer the candidate market that existed briefly. And companies know it.

Hybrid Creep: The Slow-Motion Retreat from Flexibility
The playbook is predictable. It starts with encouraging people to return to the office for two days a week, which seems reasonable. Then it moves to “three days required,” usually announced in a carefully worded email about rebuilding our culture.

Read more

There is a famous photograph of General George S. Patton taken somewhere in France in the summer of 1944. He is standing in a field, jaw set, holding a whip, looking at something just beyond the fra…

Read more

As we navigate the complexities of this “brittle” 2026 labor market and economy and the impact of AI, staying informed is more critical than ever.

To celebrate the season and our growing community, I…

Read more

The official U.S. unemployment rate stood at 4.4% as of December 2025, the highest level since 2021, but still within what economists would consider a healthy labor market.

It is a number that feels almost absurd to anyone who has spent the past year watching their LinkedIn feed fill with “Open to Work” banners, watching peers go six months between interviews, or watching entire teams disappear from companies that are still posting record profits.

The disconnect is real and measurable. The official unemployment rate tells you how many people don’t have a job. It tells you nothing about how few jobs are being created or how frozen hiring has become. To understand the 2026 labor market, you have to look beyond the headline to the underlying data.

Start here: the U.S. economy added approximately 584,000 jobs in all of 2025. That is roughly 50,000 per month, which is the smallest annual job gain since COVID and the slimmest yearly increase since 2003. Meanwhile, employers announced 1.2 million planned job cuts in 2025, up 58% from 2024 and the highest annual total since the pandemic. Long-term unemployment is at a four-year high. Planned new hires announced by employers were down 35% from 2024, the lowest level since 2011.

This is not a recession by the traditional definition. But for white-collar workers, particularly in tech, software, marketing, and customer-facing corporate roles, the conditions are indistinguishable from one.

The Layoff Wave
The most visible driver is a wave of corporate downsizing led by some of the world's most profitable companies. Amazon eliminated approximately 30,000 corporate roles since October 2025, or about 9% of its corporate workforce. It linked those cuts publicly to AI-driven efficiency and management delayering.

Salesforce cut more than 5,000 positions across two rounds, with its CEO explicitly attributing the reductions to AI agents taking over customer support and sales functions. Meta, Block, HP, Microsoft, and Pinterest have all announced significant cuts in recent months, citing AI investment and efficiency as the rationale.

The tech sector alone announced 154,000 job cuts in 2025, up 15% from 2024. October 2025 was the worst single month for announced layoffs in more than two decades. The ripple effects have spread well beyond tech: UPS announced 30,000 cuts partly in response to reduced Amazon shipping volume. The cascade is real.

The AI Attribution Problem
Here is where the narrative gets more complicated. AI is unquestionably an active force in the current labor market. Consulting firm Challenger, Gray & Christmas tracked nearly 55,000 U.S. job cuts in 2025 that companies explicitly attributed to artificial intelligence, a figure that will grow in 2026.

Read more

There is a rebellion brewing in the TA world. It started as a whisper in Slack channels and private roundtables, but last week, it spilled out into the open. A viral discussion on Reddit and LinkedIn raised a question I had only heard from a few leaders a couple of years ago.

“Is it time to drop LinkedIn Recruiter?”

For the better part of a decade, LinkedIn has been the king as the default operating system for sourcing. If you didn’t have a seat, you weren’t really in the game. It was our phone book, our background checker, and our primary communication channel.

But the mood has shifted. We are seeing the early stages of a “Tech Stack Rebellion,” driven by a phenomenon I call tool fatigue. The industry is waking up to a hard reality: paying a premium for access to a database that everyone else also has and that candidates are increasingly ignoring is no longer a strategy. It’s a tax.

Here is why the rebellion is happening, and more importantly, where the smart money is moving their sourcing strategies.

The Mathematics of Fatigue

Why are teams canceling licenses? It isn’t just budget tightening; it is a collapse in the signal-to-noise ratio.

We are currently witnessing a form of deadlock in digital communication.

1. The Automation Trap: Recruiters, armed with AI co-pilots and scheduling tools, are blasting out thousands of templated messages.

2. The “Workslop” Response: Candidates, overwhelmed by the spam, are using their own AI agents to filter inboxes or auto-apply to roles they haven’t read.

The result is silence. Response rates on traditional platforms have plummeted to single digits. When everyone has a megaphone, no one can hear anything.

Furthermore, as I have written about many times, if every recruiter uses the exact same tool to search the exact same keywords, we are all fishing in the same barrel. We aren’t sourcing; we are racing. And in a race where AI levels the playing field for speed, the only differentiator left is relationships, which a keyword search cannot find or buy.

The Shift: From “Access” to “Orchestration”

Read more