Workday gave us plenty to chew over on October 13th, after promising an event packed with “real rollouts, real results, and real ROI.” We got Agent-ready tools, Developer Agent, Custom Agents, Agent Harness, and Workday Data Cloud all in GA, while Sana became the new Workday Home.
That's a decent delivery, although the headline Talent Acquisition autopilot remains in early access, with wider availability planned for 2027.
During the Innovation Keynote, Gerrit Kazmaier argued that
“Real AI transformation is about driving growth faster, better. That's where you're really winning.”
That’s a nice quote, but we still need to know how much of that speedy growth comes from agents actually finishing work.
The bar was high before the keynote, with more than 5,500 customers using Workday-built agents, and the company also said AI SKUs generated nearly $600 million in annual recurring revenue in Q2. Yet fewer than 900 customers purchased an agent over the previous year, while around 300 had signed up for Flex Credits by Q2.
Workday further claims customers have automated 95% of frontline hiring and reduced time-to-hire by 70%. We'd be more impressed if those figures came with named customers and clearer baselines.
Workday has shipped some serious infrastructure. It still needs to prove what that infrastructure delivers when real people, deadlines, and budgets get involved.
What Did Workday Announce at Workday Rising 2026?
Workday came to Vegas with quite a hefty list of promises, and on October 13, it delivered several of them. Agent-Ready Tools, Developer Agent, Custom Agents, Agent Harness, and Workday Data Cloud all reached general availability. Sana became the new Workday Home, giving employees somewhere to see outstanding approvals, unfinished tasks, and what their agents have been doing.
Workday used a recruiter with 25 open positions to show what its Talent Acquisition Agent could handle. The agent keeps track of the vacancies that aren't moving, looks for suitable candidates, and helps with outreach and interview scheduling. Recruiters aren't handing over complete control, though. They still approve the messages before they're sent.
Joel Hellermark, Workday's chief AI officer, said:
“Our agents work like autopilots. They get the work done on their own, and bring people in exactly when their judgment makes the difference.”
Unfortunately, that hiring autopilot is still in early access, with GA planned for early 2027. The new payroll, deployment, and accountant autopilots won't reach early adopters until 2027 either.
Workday also introduced Work Intelligence, which breaks jobs into individual tasks to help businesses decide what AI should handle. It's in early access, while Decision Intelligence is already available to help teams weigh workforce decisions against financial plans.
Add the Anthropic partnership and planned C1.ai acquisition, and Workday clearly has its sights set high when it comes to work management.
Which Workday AI Agents Are Generally Available After Rising 2026?
I partially expected more specific AI agents from Rising, since several Workday AI agents were already commercially available before Vegas. Recruiting Agent, Talent Mobility Agent, Frontline Agent, Payroll Agent, BP Optimize Agent, and Self-Service Agent were all listed as GA, alongside Agent System of Record. Job Architecture Agent and Performance Management Agent remained Early Adopter, while Case Agent was still marked “Coming Soon.”
Workday did deliver a lot more on the developer and governance side. Developer Agent and Agent-Ready Tools reached GA on October 13, alongside Custom Agents, Agent Harness and Workday Data Cloud. That’s pretty handy for employers building agents that need to access sensitive people data and perform actions under company rules, although Agent Passport isn’t deemed GA yet.
Gabe Monroy’s Analyst Day demonstration was interesting too, showing an unofficial MCP connection letting an agent move an employee without triggering the required approval process. When Monroy repeated the action using Workday's newly available Agent-Ready Tools, the approval workflow kicked in, which backed up his astute claim of “Headless can’t be lawless.”
I think that’ll become an interesting topic worth discussing as more companies look for ways to enable AI agents to work across platforms and tools, alongside human staff.
Did Workday Show Agentic HR Delivering Real Business Results?
Workday had some decent evidence to bring to Vegas already. In September, Gartner named it a Leader in its first Workforce Management Technology Magic Quadrant. Early adopters of its Workforce Management Agent also reported up to 90% less time managing shift changes, 65% better process automation, and 75% fewer manual time-entry errors.
We all love numbers like that, but it’s worth saying Workday still hasn’t named the customers behind each result. Fortunately, there are stronger customer examples from its established workforce tools. Valvoline cut weekly schedule preparation from over two hours to roughly 15 minutes. Life Time reported 50,000 manager hours saved annually, $1 million in labor savings and 50% less overtime. Excellent results, although they came from Workday's scheduling suite, rather than the newer agent.
Then there's Accenture. Its October 13 Rising session described scaling more than 30 Workday agents across a 700,000-plus workforce. Bhusri also confirmed a Self-Service Agent rollout targeting 400,000 employees. Sanofi's session, meanwhile, tackled the awkward preparatory work involving legacy data and job architecture.
There’s an employee-side test, too. Workday and DailyPay found 45% of frontline workers rank scheduling among their top three reasons for taking their current job. If agents are improving workforce management, I’d eventually expect employee control over schedules to improve alongside the manager efficiency figures. Adoption numbers can’t answer that one.
Can Work Intelligence Improve Staffing and Skills Decisions?
That’s the big analytics test. Faster tasks are useful, but for workforce planning AI, the better evidence is what changes in labor cost, workforce capacity, accuracy or employee experience.
Gartner gave us a useful independent benchmark at HR Symposium/Xpo in London on October 6. It said almost 40% of companies are over budget on AI, while organizations redesigning work to change the cost of work are more than twice as likely to beat their cost goals. Still, it’s worth wondering where the capacity goes after AI makes it available.
Workday's October 5 workforce report adds another complication, showing that across more than 550 employers, demand for basic AI skills fell 25%, while demand for hands-on AI development and automation rose 51%. Internal movement also declined at 57% of employers. So who's getting retrained, and which roles are getting harder to fill?
Rising’s introduction of Work Intelligence might help. It breaks jobs into individual tasks and maps the skills behind them, helping employers decide which activities AI could handle. Workday has even said Work Intelligence can also measure how employees' time shifts as AI changes their tasks, which skills grow and whether people move into new roles. Workday Learning powered by Sana can then recommend training for those gaps, too.
Nikki Helmer, Workday's VP of Talent Management and Insights, said:
“The basic building block of work is shifting from the job to the task.”
Skills Cloud already serves more than 3,450 customers, but Work Intelligence is in early access. A task map can show a potential capacity gap; it doesn’t prove an employer has done anything about it.
Can Workday Turn Better Analysis Into Better Workforce Plans?
Workday has a more substantial planning story than another count of completed agent actions. Its October 13 expansion of Decision Intelligence gives HR and finance teams a shared view of hiring, internal mobility and budget scenarios. Managers can compare whether to recruit, develop existing employees or redirect capacity, using financial assumptions alongside the people data.
Unity Technologies offered a named example at Rising. James Choi, its global head of finance transformation and procurement, said Decision Intelligence was helping investigate revenue variances and prepare the monthly post-close outlook deck for the CFO: “It is already helping us close the gap between analysis and execution.” That identifies a real financial process. There’s still no public figure for time saved, greater accuracy, or a better financial outcome.
Workday’s existing analytics products have more concrete benchmarks. Adaptive from Workday models assumptions across revenue, costs, and workforce plans; TX Group says it cut a rolling forecast from two to three days to four hours. That’s a strong speed comparison. Workday also previewed an Adaptive Planning MCP Server intended to give supported agents permission-aware access to planning data.
We still need more information about what happens after a forecast changes. As BetterWorks pointed out on October 5, 84.8% of HR leaders surveyed were confident performance management improved business priorities, but only 43.9% could support that confidence with clear business metrics. Workday's planning tools will have to help managers explain the result of a decision, not simply produce the analysis faster.
