For HR and L&D leaders

Why it matters

Training works best when it starts from what people can already do. Without that, the budget goes to general courses for people who don't need them, while the real gaps stay hidden, and so do the colleagues who could close them. Worries about AI shape adoption too, and they rarely come up in a team meeting.

• 47%of CHROs say they haven't yet set up clear measures of AI productivityCHRO Association survey, 2026
• 36%feel they've had adequate upskilling, though 72% say the skills their job needs have changedBCG AI at Work, 2026

Questions it answers

For CHROs, People teams and L&D: the questions to settle before a training program, and after it, and where the report answers each one.

• Which AI skills do we have, and in whom?

  A directory of 64 capabilities and 125 tools and models: who can do what, what nobody covers, and what rests on a single person.

  AI Skills Distribution

• Who can teach everyone else?

  People choose what colleagues may ask them about, and only from what they rated Independent or Advanced. The 360° adds AI champions, tagged with what they're strong at.

  AI Skills Distribution · 360° AI Maturity Assessment

• Where should training start?

  Each person's three biggest gaps and three next steps. For n8n, a quiz sets the level, and where people rate themselves above their score, the report marks a blind spot: training starts there.

  360° AI Maturity Assessment · n8n Skills Assessment

• What are people worried about?

  A validated 21-statement research scale measures concern in four areas: learning, job impact, AI safety and control, and human-like AI, with general guidance for each.

  AI Anxiety Scale (AIAS)

• Why don't people use the tools we rolled out?

  The barriers people agree hold them back, including the ones training can fix, and the one fix each person asks for.

  AI Adoption Barriers

• Did the training work?

  Run it again. AI Skills Distribution matches people and capabilities across runs: moved up, unchanged or moved down. The skills assessments track the average score and the share at Practitioner or above.

  AI Skills Distribution · n8n Skills Assessment

Where to start

1. Map skills before you plan training with AI Skills Distribution: about 6 minutes per person with the default questions.
2. Hear the worries with the AI Anxiety Scale: about 5 minutes, anonymous by default. Run it before a rollout to set a baseline, and again after a big change.
3. Check tool skills with a quiz. The n8n Skills Assessment takes about 20 minutes; the Claude Skills Assessment, for everyone who uses Claude at work, is coming soon.
4. Run them again after training to see what moved.

Running an AI-literacy program, for example for Article 4 of the EU AI Act? A run before and after training gives you a baseline for role-specific training and a record of progress. It's evidence for your program, not a compliance certificate.

Who sees what

Each survey has its own defaults, and whoever launches it can change them:

• AI Anxiety Scale: anonymous by default. People don't see their own result.
• AI Adoption Barriers: anonymous with the one-click launch.
• 360° AI Maturity Assessment: names people by default, through Google sign-in on your company's domain, because the person and team views need to know who answered. It can run anonymously instead.
• AI Skills Distribution: names people, because it's a directory of who can help. The survey says so on its first screen.
• Skills assessments: anonymous with the one-click launch. People see their own level right after answering, unless that's switched off, but never the right answers.

No survey ranks people, and the results are for development, not performance reviews. There are no integrations or tracking agents: nothing reads people's tools, files or messages. Results show from the first answer, with no minimum group size, so in a small team they can point to one person.

What it won't tell you

Proven proficiency. Using a tool is a useful directory fact, not evidence of expertise, and the levels in AI Skills Distribution aren't validated proficiency. The quiz-based assessments go further, but three questions per area is a quick check, not an exam.

A diagnosis. The AI Anxiety Scale measures the concern people express; it can't say why, and it gives no mental-health advice. It's in English only for now.

A benchmark. Averages are your organization's own; there's no external comparison.