AI now sits in almost every stage of talent management, on both sides of the table. Role briefs, CVs, screening, interview notes and development plans are all faster to produce. The trouble is that not all of it is reliable, and unreliable inputs lead to real consequences: the best candidate screened out before a human sees them, development budget spent on the wrong people, and genuine potential overlooked.
In this on-demand session, Chief Commercial Officer Hannah Mullaney is joined by Chief Technology Officer Barny Ritchley and Principal People Scientist Jake Smith. They show where unproven AI distorts hiring and development decisions, why small errors multiply as they move through the funnel, and which signals still hold up.
One experiment discussed in the session, run by AI and HR compliance specialist Martyn Redstone, put the same batch of CVs through three AI models every day for three weeks. One CV ranked first, then dropped out of the top 10 the next day. On average, more than half of the CVs were never read before a shortlist came back.
Plus, Jake shares our latest data on whether candidate AI use is shifting assessment scores, and the safeguards that flag it when it does.