Why LinkedIn's Ryan Roslansky Says AI Now Sets the Pace at Work
Degrees are losing their hiring pull. Ryan Roslansky, LinkedIn's CEO, says the advantage goes to people who are adaptable, forward-thinking, and ready to learn with AI-regardless of pedigree.
The data backs the pressure on graduates: unemployment for recent grads sits at 4.8% vs. 4% for all workers, per the New York Fed. See the trend here.
The new hiring signal: AI fluency over pedigree
Roslansky's take is blunt: the future of work won't favor the "fanciest degrees." The signal executives should hire for is practical AI competence plus a willingness to learn.
He frames AI as "a second brain" that knows him well-and he uses Copilot to draft nearly every high-stakes email to CEOs and world leaders. That's a clear cue for how leaders should work: set standards by using the tech yourself.
Degrees are losing leverage
This isn't new in tech. Mark Zuckerberg, Sam Altman, and Jack Dorsey built careers without degrees. Even Zuckerberg has questioned whether college is preparing people for the jobs that exist now, especially given student debt.
The takeaway for executives: stop over-indexing on diplomas. Shift your screening to skill assessments, portfolios, and AI-first workflows.
A shifting market for talent discovery
OpenAI is moving into hiring with an AI-driven jobs and certification platform to match companies with AI-savvy workers. The goal: help firms find people who can apply AI to real tasks and verify capability with credible signals.
Whether you adopt that platform or not, the direction is clear-competence proof will outpace credentials.
What leaders should do next
- Rewrite role requirements: list the core workflows candidates must automate or accelerate with AI, not just years of experience.
- Screen with scenarios: give take-home tasks that require AI tools (prompting, data analysis, summarization, coding support) and assess outputs, not process theater.
- Require AI use in leadership comms: adopt Copilot-class tools for board updates, investor notes, client proposals, and partner outreach.
- Stand up an internal AI guild: a cross-functional team that publishes prompts, playbooks, and ROI wins every month.
- Fund learning sprints: 4-6 week upskilling cycles tied to KPIs-pipeline velocity, cycle time, support resolution, code throughput, research turnaround.
Operating metrics to track
- Time saved per workflow automated (baseline vs. post-AI)
- Quality uplift (error rate, win rate, NPS, CSAT)
- Throughput per head (tickets closed, features shipped, proposals sent)
- Cost per deliverable (designs, analyses, campaigns)
- Model/tool usage vs. outcome correlation (not vanity logins)
Policy and guardrails (without slowing the work)
- Data classification: what can and cannot be used with external models
- Prompt hygiene: remove PII, sensitive client info, and sealed financials
- Attribution: declare AI assistance in research-heavy or legal-adjacent work
- Human review: define sign-off thresholds for public or regulated outputs
Skill paths and quick wins for your org
- Executive stack: use AI for weekly updates, speech drafts, market scans, and board docs. Standardize a prompt library for your leadership team.
- Sales: AI-generated account research, first-draft outreach, objection handling, and deal reviews with call summaries.
- Ops/Finance: reconciliations, variance analysis, vendor comparisons, and policy drafts with clear audit trails.
- Product/Engineering: PRD drafting, test generation, bug triage, and code review assistance with clear boundaries.
- HR/Talent: skills-based assessments, structured interview scorecards, and onboarding guides built with AI.
Where to upskill fast
If you need structured, role-based learning and certifications your team can complete in weeks, explore:
- AI courses by job role for execs, product, sales, marketing, and ops
- Practical AI certifications focused on automation, ChatGPT, Claude, data analysis, and coding
The executive mandate
Hiring for degrees is an outdated filter. Hire and develop for AI fluency, curiosity, and speed of learning.
Lead by example, measure outcomes, and turn AI from an experiment into a performance system. The teams that learn fastest will win this cycle.
Your membership also unlocks:
AI Capex on the Hot Seat: Apollo Exec's No Comment on Vendor Financing and Capex Recycling Stirs Transparency Debate