Talent Intelligence help us understand the skills and experience of people in an organisation. With good intelligence about talent we can make better decisions, reduce costs, help individuals and ready ourselves for the changes and crises ahead.
This article is about the future of talent intelligence using AI (artificial intelligence). It includes aspirations, fears and how to get started.
Table of Contents
Talent Intelligence – the transformation starts
Our far distant ancestors recognised employee’s talents. Talent was assessed by humans, sometimes with simple measures. But there’s a limit to how much humans can track, and humans have prejudices, and our knowledge is limited.
Since the late 1970’s there have been commercial tools that record aspects of talent for recruitment boards. (I got my first job using one of them.) Since then there has been decades of academic research into how to improve them.
From 2024 a new revolution began in talent intelligence. It was driven by AI, with its ability to process vast quantities of data. Now we can build learning and work environments that are continually monitor skills and aptitudes, collate the data, and analyse it. The result is a vastly richer understanding of talent.
Talent intelligence has the potential to help professionals to accelerate their career development. It can help managers reduce costs and manage successions. And at an operational level, it can help us manage disruptive changes. Ironically, that it includes reorganisations caused by the introduction of other types of AI.
It’s summer 2025 and talent intelligence is being transformed.
But into what?
The Talent Intelligence dream – 4 aspirations
Empower careers
Context. Career paths give professionals a vision and a route to achieve it. The route involves building hard skills and soft skills, and some green skills, and practicing the all.
When hiring people, or moving them to different roles, we look at their skills and the roles they have taken. The challenge is to do this efficiently and fairly when there’s so limited information available. It’s also become urgent. There are many areas with chronic skill shortages, and fast staff turnover caused by employees hunting for better opportunities elsewhere.
The aspiration for managers is to recognise people’s potential, have the ability to develop it, and use that to retain key staff. This requires intelligence about the individuals, the company’s needs and the trends elsewhere. That’s hard for humans to achieve. However combinations of AI engines can bring that information together.
Adopt AI efficiently
Context. AI is a “disruptive” technology that changes the way people work. It allows us to achieve much more than before. Thanks to AI, we increasingly focus on the complex aspects of work. To flourish in an AI world, we use more of the human “soft” skills. We also draw on hard skills and experiences that are not listed on our CVs.
For employers watching us, they barely know what skills we’re now using, and they guess at what skills we’ll need in the future.
The aspiration is to have the right skills in place as we transition to AI. That requires talent intelligence to discover and map the talent we already have. It requires monitoring of talent trends as AI is introduced. And companies need to be able to rapidly develop the skills in their talent pool.
Discover what people are really doing – reorganise safely
Context. Companies and major institutions have knowledge of their finances, often down to a fine detail. They have a myriad of measures around their products and services, and their productivity and quality, and their suppliers. And with market research and careful monitoring, organisations can learn much about their customers. But of their own staff, there’s much less. There’s some information around achievements, but very little around their talent. (How often do HR managers have to use LinkedIn.com to learn about their own staff?)
Without talent intelligence, it’s massively risky to reorganise – unexpected gaps appear. And it’s worse for companies with complex, supply chains, manufacturing processes or distribution networks – even tweaks to them can be catastrophic.
The aspiration is ultimately to understand the full detail of what people are doing, and the varied talents they use to achieve this. That requires measurement, collation and reporting.
Handle sudden crises
Context. We face crippling cyber attacks, disruptions to supply chains, natural disasters, wars, and sudden taxation changes. They can cost lives, cause human misery and devastate companies. It might not be possible to avoid them, but we can prepare for them.
The aspiration. When crises hit, there’s been sufficient contingency planning, and everyone who matters is well practiced – that requires talent development, which in turn requires talent intelligence. So when surprises occur during the crisis – which they will – there are people who have the talent to make innovative responses.
N.B. For an extensive description of managing crises, see David Omand’s “How to Survive a Crisis”, Penguin, ISBN 9780241995402, https://www.penguin.co.uk/books/448107/how-to-survive-a-crisis-by-omand-david/9780241995402
Dreams and aspirations are good, but things always go wrong …
Talent Intelligence’s future – 4 dangers
As of summer 2025, the vast majority of people’s experiences of AI is for specific “point solutions”. The solutions solve a specific problem. If the tech fails, we swap to another one. That’s currently easy, but once AI is embedded it will become harder. With that in mind, we have to prepare for the things that can go wrong.
1. The risk of misjudgements about individuals
Context. There’s a danger of bias from talent assessments and recommendations. Ethical AI companies work to use models that are neutral towards gender, beliefs, ethnicity (and more). There are also subtle biases from cultural differences, which need to be addressed.
There is a risk that AI could misjudge individuals, resulting in discrimination from AI engines that were meant to be neutral. Initially humans assess the results of AI, but as it becomes more reliable so the scrutiny becomes less. Ultimately AI could be deciding people’s futures.
(I’m focussing on neutrality as an objective. There’s a counter-argument that AI should bias in favour of groups that need special support. That’s a much broader moral and political discussion.)
Mitigation. Legislation may help, but it can take years and the technology has often changed by that point. The quickest response is for industry to create norms of “good practice”, train their staff and monitor what happens. And it’s important to build trust, by being open about problems and what is being done to resolve them.
2. Talent Intelligence AI could delay the take-up of AI itself
Context. Talent intelligence uses models of careers for professionals. Practices within the professions evolve to support new insights, technologies and ways of working. However AI is trained on data that is readily available, and often it is years old.
The danger is that AI hampers organisation’s ability to adapt to emerging new needs. (Ironically, that means AI makes it more difficult to embrace AI.)
The mitigation is to track the AI’s strengths and weaknesses, and compensate where needed. Obviously that requires human expertise and “common sense”.
3. Career paths can become a straightjacket
Context. In professional disciplines, job are increasingly based on established roles. To get a good job, your CV and self-presentation is expected to excel at a single role. If you’ve got deep expertise at two roles then your CV will include material that is irrelevant to your job application.
The danger is that AI-based talent intelligence will enforce career path rigidity. So if you want to change career, the AI will discriminate against you. And if you’ve had a long and diverse career, the AI may also discriminate against you.
AI systems can be designed to avoid this. For example, talent intelligence engines can assess people’s potential, then plan for their retraining. But will all AI suppliers do that?
And AI systems could also help you redesign your CV and profiles to focus a single role.
4. New entry barriers to the job market
Context. An AI-based assessment of job applications uses the candidates CVs, compares the CV against job description and uses smart techniques to interpret subtleties behind their wording. The AI uses a much broader experience base than humans. However the CVs of students focus on academic detail. Many students have little or no experience of work, or equivalent roles within sports, team hobbies and charities.
The danger is that newcomers to the job market can’t find an entry, and poorly implemented talent intelligence exaggerate this. Students who focus on “cramming” for academic results, could suffer most.
The mitigation is both technical and practical. Technically, we can build AI-based assessments that explore a candidates problem-solving capabilities, cooperation skills, communication abilities, and other soft skills. That can provide indicators of some of the skills employers are looking for. And AI can suggest up-training courses that could further develop a potentially good candidate.
And practically, employers need to massively increase the number of internships so work newcomers can put experience on their CVs. (A political point: companies need incentives for this, because managing internship costs money, and only a few of their interns will migrate to become their employees.)
“Talent Intelligence is being reinvented.
To achieve that, everything can be questioned.”
Talent Intelligence – a path forwards
Achieving an aspiration takes more than dreams and fine words. It takes hard work.
For each of the tactics listed below, there’s a common sequence:
– research what is likely to be needed in the medium term,
– conduct trials of existing tools, to build expertise,
– map the gap between your current and future needs,
– then build a plan for getting there.
1. Start collecting data
For deep intelligence about talent we need intelligence collection – data about roles, skills, training courses and especially people.
To achieve that, companies need to use online platforms and integrated tools that can measure everything of potential value. For some areas, new collection tools are needed, such as for assessing people’s skills.
Collectively, the tools need to convert their raw data into meaningful metrics. Then collate that data into repositories. These can then be interpreted and analysed by both AI and humans. Then that needs to be presented to the decision-makers. AI can help them, too.
There’s a practical implication: Companies need to invest in technology to capture and store this information.
2. Understand and adapt the models
Talent Intelligence relies on theoretical models about the different aspects of talent, and how they relate to roles and skills, and more. There’s models about how people learn and adapt. And there’s models about how hard, soft and green skills can be classified, subclassified, and measured.
Talent Intelligence tools and AI rely heavily on these models.
As part of developing your organisation’s own “future of work”, you’ll find yourself studying the models, to understand their strengths and weaknesses. If you’re in a large or specialist organisation, you may want to adjust models to suit your own way of working.
As with data collection, that involves a combination of research, trials and strategic analysis.
3. Invest in tool integration
The move towards a talent intelligence future will involve using different tools. Of these, many (or most) will rely on AI. Tools include data collection and assessments, collation, analysis, presentation and reporting. They also include talent development (online learning) for your staff.
For the IT development teams, these tools need to be interfaced with data pools, to add to the pools or use them. There are significant issues for data integrity, data security and business continuity that come from this.
For large enterprises, this will require a careful strategy. If you wait to learn from what other companies do, they’ll have stolen the advantage from you. But if you build a “final solution” without trials, you could live to regret it. That applies to introducing AI in general, and specifically to talent intelligence.
The conclusion? Start now, with trials.
4. Rethink talent management
Tomorrow’s talent Intelligence will involve a new ecosystem of intense data collection, increasingly smart AI, and a library of online tools for developing and managing talent.
With that comes changes in practice, including shifts of responsibility from HR to line managers, new challenges for HR teams, an increased focus on developing employees careers and skills, new reporting to company executives, and much higher expectations.
Talent intelligence – a path forwards
They say that the future comes at its own speed. That’s misleading. It’s possible to accelerate change with careful planning and the right processes. That requires change management tools, including training, research, analysis, planning, development and support.
But first, you need to know where you are heading.
Caveat.
The opinions in this article are mine, and are influenced by my work in documenting aspects of the intelligence profession – here. The comments do not necessarily represent those of Metatalent.ai Ltd – a company I co-founded. (Metatalent.ai provides AI technologies and new methods for talent intelligence, supporting talent acquisition, talent development and talent management. The company also provides a range of professional services to help with the transition to the future of work.)
— Adrian Cowderoy, May 2025
