The org chart is holding back your A.I. strategy

Ryan Roslansky and Aneesh Raman did not set out to write a technology book. Open to Work: How to Get Ahead in the Age of AI, published today by HarperCollins, is explicitly a book about humans - what makes them irreplaceable, what is holding them back, and why the organisations that employ them may be a bigger obstacle to A.I. productivity than the workers themselves.

The message from Roslansky, CEO of LinkedIn, and Raman, LinkedIn's chief economic opportunity officer, is direct: the transition to an A.I.-augmented workplace will be worker-led, not top-down. For Australian HR professionals overseeing transformation programs built around structured rollouts and mandated training, that framing presents a genuine challenge.

The org chart problem

The most confronting argument in the book - and the one most directly relevant to HR leaders - concerns the fundamental structure of how organisations are designed.

"The org chart was built in the industrial age to bring order, predictability, and stability to rapidly growing organisations," Raman says. "Companies need to let that go, as it's going to hold back innovation."

Raman's argument is not that hierarchy is inherently wrong, but that the industrial logic of the org chart - designed for efficiency, scale, and predictability - is precisely the wrong architecture for capturing the value A.I. makes possible. Where productivity gains from previous technology waves came from doing the same things faster, A.I. creates value through genuine novelty: solving problems in ways that have not existed before, crossing the departmental lines that org charts are designed to enforce.

"Where you're going to see the real returns on AI isn't just a new workflow around AI, but rather new work around human capability," Raman says.

For Australian HR leaders, that distinction carries significant operational weight. Many enterprise A.I. programs have been designed as efficiency plays - automating existing processes, reducing headcount in defined functions, accelerating tasks that were already being done. The LinkedIn executives argue this misses the larger opportunity, and that realising it requires a tolerance for worker-led experimentation that most large Australian organisations have not yet developed.

The skills that survive

Drawing on conversations with neuroscientists, organisational psychologists, behavioural economists, and talent leaders, Roslansky and Raman identify five human capabilities - what they call the 5Cs - that they argue A.I. cannot replace and that employees and employers should be actively developing.

The five are curiosity, courage, creativity, compassion, and communication.

The framing is deliberately non-technical. "AI can generate possibilities based on patterns. Humans decide which ones matter," the authors write of curiosity. On courage: "AI can calculate risk. Only humans decide what risk is worth taking." On creativity: "AI can remix what exists. Humans decide what's worth reimagining." On compassion: "AI can simulate concern. Only humans feel it and express it." On communication: "AI can translate language. Only humans can turn language into meaning."

The 5Cs framework has immediate implications for how Australian organisations approach capability development. Skills frameworks built around technical competencies - coding, data analysis, digital literacy - are necessary but insufficient if the adjacent human capabilities are not being developed alongside them. The risk, as the book frames it, is building a workforce that can use A.I. tools but lacks the judgment, courage, and creativity to use them well.

The data behind the urgency

The book is grounded in LinkedIn's global labour market data, and the numbers the authors cite should focus the attention of every Australian people leader.

Nearly 90 per cent of C-suite leaders globally say accelerating A.I. adoption is critical - not eventually, but now. Two-thirds of corporate leaders say they will not consider job candidates without A.I. skills. LinkedIn data shows that 24 per cent of the skills required for the average job changed between 2015 and 2022. The authors estimate that by 2030, as A.I. accelerates that pace of change, the proportion of skills that will have changed could reach 70 per cent.

Read next: Consistency is key: HR leaders share trust-building strategies at National HR Summit

That last number deserves to sit with Australian HR leaders for a moment. A workforce planning horizon of five years - which most Australian enterprise workforce plans do not extend much beyond - may now be looking at a world where seven in ten skills in a given role have materially changed. The capability development implications are significant, the workforce planning implications are profound, and the time available to respond is shorter than most organisations have yet acknowledged.

Read next: Too much AI causing 'brain fry' among employees

By early 2025, A.I. literacy had become one of the skills LinkedIn members from around the world added most to their profiles. As early as 2023, one-third of content writers on the platform had added A.I. literacy skills - significantly outpacing software engineers at 19 per cent. Graphic designers and marketing managers showed similar momentum.

Worker-led, not top-down

The book's most practically disruptive argument for HR departments is its insistence that the A.I. transition will not be driven by enterprise transformation programs. It will be driven by individual workers experimenting in their day-to-day roles, often without permission and across the boundaries of their job descriptions.

"It's going to be a worker-led transition, and so companies are going to have to figure out how to let individuals start to move into this new era in their day-to-day work," Raman says. "We have more autonomy than we often think in terms of pushing for what we want to do that might push our work to the next level.”

Read next: GenAI to hit women's jobs more than men's, ILO warns

Raman frames this through three categories of activity. The first covers things A.I. already does well today - generating code, running quick analyses, producing first drafts. The second involves experiments to create something genuinely new with A.I. The third, and most valuable, involves using the time freed from the first category and the lessons from the second to start using A.I. collaboratively. "What are you doing with other people?" he asks.

For Australian HR professionals, this worker-led framing creates a governance question that most enterprise A.I. policies have not yet resolved. If adoption is fastest among workers who experiment independently - crossing team lines, testing tools outside approved stacks, adapting processes without waiting for sign-off - how does the organisation create the conditions for that experimentation while managing the data, privacy, and compliance risks that accompany it? This is particularly acute in Australia, where Privacy Act obligations and the emerging regulatory environment around automated decision-making create specific constraints that need to be built into A.I. governance from the outset.

The compassionate case for urgency

Roslansky and Raman are explicit that the anxiety workers feel about A.I. is legitimate and understandable. "Your resistance is biology, not weakness," they write. "That knot in your stomach when you think about AI? That's millions of years of evolution trying to protect you from rapid change."

But the book does not use that acknowledgement as a reason to slow down. Raman's answer to workers who would prefer to maintain stable responsibilities without engaging with A.I. is blunt: "Nobody is coming to save any individual but themselves."

The tone is sympathetic but unsparing. "There was a career ladder, and there was extreme clarity about what you had to do to get on each rung of that ladder," Raman acknowledges. But he is ultimately optimistic about the direction of change: "Very few people have ever had real control over their career. Because of AI, I think we're about to have the first generations at work that have more control over their career than any who've come before."

For HR leaders managing the cultural dimension of A.I. transformation - communicating change, managing anxiety, building the psychological safety required for genuine experimentation - the book's emotional register offers a useful template. Acknowledge the fear. Name the biology. But do not use it as an excuse for delay.

What it means for Australian HR

The practical implications of Open to Work for Australian HR professionals cluster around four areas.

The first is skills architecture. If 70 per cent of role skills may change by 2030, the competency frameworks that underpin most Australian performance management and development systems need urgent review. The 5Cs framework offers one starting point for identifying what to preserve and develop alongside technical A.I. skills.

Read next: AI ‘beginner tax’ costing Australia billions, as workers left to ‘shadow AI’ learning

The second is governance for worker-led experimentation. The organisations that capture the most value from A.I. will be those that create safe, structured space for employees to experiment with tools and processes, including outside their immediate job descriptions. Most Australian enterprise A.I. governance frameworks are not currently built for this.

The third is the org chart question itself. Raman's argument that rigid hierarchical structures will hold back A.I. adoption has direct implications for how HR leaders think about team design, role definition, and the degree of cross-functional latitude employees are afforded in practice, not just in policy.

The fourth is the skills-first hiring shift. LinkedIn's pivot toward skills-based rather than credential-based hiring - assessing what candidates can do rather than where they have been - is accelerating in line with A.I. adoption. For Australian talent acquisition teams still relying heavily on degree qualifications and job title histories as proxies for capability, that shift requires a deliberate strategic response.

The world’s first Org. Chart