Phenom’s AI Day Bets on Trust as AI Hiring Faces a Legal Reckoning

Phenom, a Philadelphia company that sells artificial intelligence software to corporate HR departments, opened registration on July 15 for AI Day, its fifth annual conference on AI in human resources. The broadcast streams live from Phenom Studios on September 17, built almost entirely around one theme: proving that AI agents can be governed, audited and trusted before HR departments hand them real hiring decisions.

That pitch lands at an awkward moment. A federal judge just let a discrimination lawsuit against rival HR vendor Workday move forward, and Phenom’s own research shows most employers still do not know what to do with the AI tools they already have.

Governance Runs Through Every Session at AI Day

Phenom calls itself the leader in applied AI for HR and says its platform is the only AI infrastructure built specifically for the function, a claim from the company’s own marketing materials rather than an outside audit. Five sessions built around that claim will run on September 17, all aimed at one audience: HR and IT leaders deciding how much of recruiting and workforce management to hand to autonomous agents.

Registration promises live demonstrations and a chance to explore the mechanics of AI built for HR, in the company’s words. The sessions themselves read like a checklist for anyone worried about handing decisions to software they cannot fully see inside:

• Architecting the AI Workforce, One Unit of Work at a Time – how the platform defines each job by industry, role, geography and workflow, then assigns the matching agent without custom setup.
• The Orchestration Layer that Governs the Work, Not Just the Speed – the infrastructure coordinating multiple AI agents, autonomous workflows and human judgment in real time.
• Elevating HR Operations to Drive Workforce Change – using task-level data to decide what gets automated and what gets upskilled instead.
• Building AI Frameworks that Earn Trust, Not Scrutiny – a live, auditable record of every AI decision and every human escalation.
• Unifying AI, Governing Every Decision – folding existing HR and AI tools into one governed data foundation with full audit logs.

“Most AI deployments fail because the technology applied was never designed to solve the specific unit of work,” said Kumar Ananthanarayana, Phenom’s vice president of product management. “At AI Day, we open up that architecture so HR and IT leaders can see exactly how context, orchestration and governance work together.”

A Federal Judge Just Made the Software the Defendant

Phenom is not building this pitch in a vacuum. Derek Mobley, a job applicant who says Workday’s screening software rejected him from dozens of jobs, sued the company in 2023.

He argues the tools downgraded him through proxies tied to his age, race and a disability rather than any explicit protected trait, claims that invoke the Age Discrimination in Employment Act (ADEA), Title VII of the Civil Rights Act and the Americans with Disabilities Act (ADA).

The case has moved quickly through 2026. A court authorized notice to a wider class of applicants in February, and in June, U.S. District Judge Rita Lin in San Francisco rejected Workday’s argument that California’s anti-discrimination law should not cover people the software screened for jobs outside the state.

The ruling treats the vendor itself, not only the employers who buy its product, as a potential target for discrimination claims.

Our technology looks only at job qualifications, not protected traits like race, age, or disability. We rigorously test our products as part of our Responsible AI program to confirm our tools do not harm protected groups.

Workday said in a statement, denying that its software makes hiring decisions “in California or anywhere else.” A second, separate lawsuit filed in California on January 20, 2026 accuses Eightfold AI of generating secret candidate “likelihood of success” scores without telling anyone, allegedly in violation of the Fair Credit Reporting Act (FCRA) and California’s Investigative Consumer Reporting Agencies Act (ICRAA). Eightfold’s platform is used by companies including Microsoft and PayPal.

Neither case has reached a final verdict yet. Both test whether a hiring algorithm’s decision can be defended in court the way a human recruiter’s can.

Recruiters Trust the Algorithm More Than Candidates Do

Survey data backs that up. A 2025 Greenhouse study of more than 4,100 respondents found 70% of hiring managers trust AI to make faster, better hiring decisions, versus just 8% of job seekers who called the process fairer for it.

A newer Greenhouse survey of nearly 3,000 candidates, published in 2026, found only 26% trust AI to evaluate them fairly. Most were not told AI was involved until they were already deep into the process.

National polling backs up the discomfort. More than eight in ten American adults voice concern about bias in AI-based hiring decisions, and roughly two-thirds say they would not want to apply for a job where AI makes the final call.

The unease is not limited to hiring, either. Similar questions about undisclosed AI data use have surfaced elsewhere, including how political campaign text bots quietly harvest voter data without clear disclosure to the people texting back.

Phenom’s Own Benchmark Admits the Market Isn’t Ready

Phenom’s inaugural State of AI and Automation for HR benchmark report, released earlier in 2026, found that 83% of organizations show low AI and automation maturity in human resources, and 30% of HR professionals say they have limited knowledge of how to apply AI in talent acquisition at all.

Jonathan Dale, Phenom’s vice president and general manager of global marketing, said the data show most organizations are still in the early stages of putting AI to work in HR.

Gartner’s numbers point the same way. A Gartner survey of 114 HR leaders in October 2025 found 88% had not yet seen significant business value from their AI tools, even though 61% described their organizations as being in advanced stages of implementation.

Part of the gap is definitional. Analysts describe an industry-wide pattern of vendors rebranding existing virtual assistants as agentic AI to capture buyer attention, without the software actually planning or acting on its own.

Gartner estimates only about 130 vendors currently offer systems that clear that bar, and projects 40% of agentic AI projects will be shelved by 2027 over cost and unclear return.

Is This Actually Agentic AI, or Just a Chatbot in a New Suit?

Mostly not yet, according to analysts who track the category closely. Gartner separates AI assistants, which respond to prompts, from AI agents, which plan and act with limited supervision. Most hiring software marketed as agentic still behaves like an assistant, handling narrow tasks such as sourcing or scheduling rather than running an entire hiring funnel on its own.

A handful of products have pushed further. Chipotle says an agent built by recruitment-software company Paradox cut its time to hire by 75% across its restaurants, and LinkedIn’s Hiring Assistant, launched in late 2024, automates large portions of sourcing and initial screening.

Workday, the company now fighting Mobley’s lawsuit, says its own Recruiting Agent handled 14 million hiring processes in a single quarter, up 44% year over year, across more than 4,000 customers running at least one of its agents.

The Compliance Deadline Nobody Can Quite Agree On

Underneath the trust gap sits a regulatory one. The European Union’s AI Act classifies employment tools, including CV screening, candidate ranking and video interview scoring, as high risk under the law’s Annex III, and once obligations take effect, employers and vendors face mandatory risk assessments, bias testing, human oversight and candidate disclosure requirements.

New York City’s Local Law 144 already requires an annual bias audit and candidate notice for any automated hiring tool.

Exactly when the EU rules bite depends on which tracker gets read.

• One 2026 industry report puts full enforcement for high-risk employment AI at August 2, 2026, matching the law’s original schedule.
• A separate 2026 HR-technology analysis says that date has since been pushed to December 2, 2027, after what it describes as a formally agreed delay.

Either way, the direction is the same. Audit trails and bias documentation are starting to outweigh feature-of-the-month wins in procurement decisions, according to one recruiting-technology research report.

Phenom’s cameras start rolling on September 17 regardless, in front of the exact HR and IT buyers who will decide, contract by contract, whether governance is an actual product or just this year’s word for trust.

Frequently Asked Questions

What is agentic AI in HR?

Agentic AI describes software that can plan, decide and act toward a goal with limited human input, rather than just responding to a prompt the way a chatbot does. Gartner projects that by 2030, half of current HR activity could be automated or handled by AI agents, though most tools sold as agentic today still run under heavy human supervision.

Is it legal to use AI to make hiring decisions?

Yes, though enforcement varies sharply by state. Several states, including Kentucky, Indiana, Ohio and Tennessee, have no AI-specific employment statute yet, and courts have been applying existing discrimination law instead, the same approach used against Workday. Employers in the EU and New York City already face specific audit and disclosure rules that most of the rest of the U.S. does not.

Are job candidates told when AI is screening them?

Not consistently, and not always in advance. Greenhouse’s 2026 survey found almost two-thirds of U.S. job seekers said they had already gone through an AI interview in the past six months, a sharp jump from prior years, even as disclosure practices lag behind adoption.

What happens next in the Workday lawsuit?

The case remains in discovery after Judge Rita Lin rejected Workday’s jurisdictional defense in June 2026. Applicants who fit the certified class had until March 7, 2026 to opt into the collective action, and the disability discrimination claim is the piece drawing the most attention from corporate HR legal teams.

When can I watch Phenom’s AI Day?

The Society for Human Resource Management (SHRM, the trade group that accredits HR training) has accredited the broadcast, which airs live from Phenom Studios on September 17 at 10 a.m. ET (4 p.m. CEST) and stays available on demand afterward through registration on Phenom’s event site.