AI for CHROs · Independent decision intelligenceSource-backed reporting · No paid editorial rankings
CHRO AI Current

A workplace-AI publication for people leaders balancing workforce capability, employee experience, operational evidence, and the legal and human consequences of algorithmic decisions.

Workforce updates

An AI HR service must recognize accommodation requests without legal terms

EEOC guidance says an applicant or employee can request a disability-related adjustment without writing the request or using the terms ‘reasonable accommodation,’ ‘Americans with Disabilities Act,’ or ‘disability.’ A CHRO approving an AI intake channel needs evidence that ordinary employee language reaches a qualified human process instead of being classified away.

Answer capsule

EEOC guidance says an applicant or employee can request a disability-related adjustment without writing the request or using the terms ‘reasonable accommodation,’ ‘Americans with Disabilities Act,’ or ‘disability.’ A CHRO approving an AI intake channel needs evidence that ordinary employee language reaches a qualified human process instead of being classified away.

What the source establishes

  • EEOC says an applicant or employee may indicate a need for an adjustment or change in the application process or at work for a reason related to a medical condition.
  • The agency says the request does not have to be written and does not have to include the terms ‘reasonable accommodation,’ ‘Americans with Disabilities Act,’ or ‘disability.’
  • EEOC advises employers not to automatically refuse a request or use an inflexible policy, and to review each request individually because there is no one-size-fits-all accommodation.
  • The guidance describes employer responsibilities at a general level; it does not validate an AI classifier, determine whether a particular request is covered, or prescribe an automated eligibility decision.

Treat recognition as an intake duty, not a keyword test

Employees may say that a schedule, workspace, communication method, application step, tool, travel requirement, or policy needs to change because of a medical condition without naming a statute or formal process. If an AI service only opens the accommodation path when it sees a controlled phrase, the channel can miss the very request it is meant to receive. The CHRO decision should define the range of ordinary-language signals that require safe routing and what happens when the system is uncertain.

The acceptance record should test written and spoken requests across the channels actually offered, including incomplete, indirect, misspelled, translated, assistive-technology, and manager-submitted examples. Include requests embedded in a longer question and cases that are not accommodation requests. Measure missed routing, inappropriate routing, delay, user abandonment, and whether the person can reach a human without repeating sensitive information. A polished answer is not evidence that the request entered the governed process.

Keep intake separate from the individual decision

Recognizing that a message may require accommodation review is not the same as deciding disability status, entitlement, effectiveness, undue hardship, job duties, or the appropriate adjustment. Those questions are individual and contextual. The AI service should not turn a confidence score into an automatic denial, promise, medical conclusion, or manager instruction. Its approved authority should be limited and visible to the employee, HR team, and process owner.

For every routed case, preserve the originating language, time, channel, consented transcript or record, system version, routing reason, recipient, acknowledgement, handoff status, human decision owner, and correction path. Limit access to the information needed for the role. The record should show when the system summarized or translated the request and allow the person to correct that representation before it shapes a consequential response.

Make the alternate path equally usable

An accommodation intake channel cannot depend on the same interaction pattern that a person is asking to change. The CHRO should require an accessible alternative for someone who cannot use the chatbot, voice interface, portal, authentication flow, timed session, language, or device. The alternative should be findable at the point of need, reach an accountable person, and preserve the original request date rather than restarting the clock after a failed automated interaction.

Test the service with people who use relevant assistive technologies and with the workforce environments in scope. Inspect keyboard and screen-reader use, captions, contrast, language access, low bandwidth, shared devices, authentication recovery, privacy in workplace settings, and escalation outside normal hours where promised. Accessibility conformance claims and generic usability scores do not show that this specific employee can submit, understand, correct, and follow the request.

Govern the employment process beyond the interface

A successful handoff still needs acknowledgement, individual review, appropriate discussion, documented alternatives, confidentiality, timely communication, and a way to raise another or changed need. Monitor cases that disappear between intake and assignment, repeat requests, reversals after escalation, and differences by channel or population. Do not optimize the service around containment or ticket deflection if those measures reward the system for suppressing a protected or sensitive request.

The EEOC page provides general guidance and links to broader accommodation resources; it does not approve an AI HR service or resolve a specific case. The final CHRO record should state the covered employer and population assumptions, applicable jurisdictions, allowed system actions, human ownership, privacy boundary, test evidence, unresolved failure modes, and review trigger. Current facts and qualified HR, employee-relations, disability-access, privacy, security, labor, and legal review control.

Turn this source into a reviewable decision

For AI for CHROs, use this briefing as a dated decision record rather than a substitute for the source. Preserve U.S. Equal Employment Opportunity Commission, the exact URL, the August 11, 2026 review date, the supported facts above, the editorial interpretation, the limitations, and any buyer-specific evidence. Link that record to the decisions most directly affected: HR policy and employee service; Workplace AI governance and change; Recruiting and candidate support; Performance and work allocation. State whether the source changes the scope, evidence requirement, control, sequence, or only the language used to describe the decision.

Before action, name the accountable owner, affected population and workflow, exact offering or configuration, source data and rights, human decision point, exception and appeal path, complete cost, expected benefit, failure and stop conditions, retained evidence, and next review date. Keep official facts, provider statements, buyer observations, representative tests, measured outcomes, editorial inferences, and unknowns visibly separate. Reopen the record when the source, offer, model, integration, data, policy, population, responsible person, or measured result changes.

Limitations and unknowns

The EEOC page is general guidance and does not have the force and effect of law, decide a specific accommodation request, validate a classifier, require a particular technical design, or establish that automated intake is appropriate. Coverage, obligations, confidentiality, timing, and effective accommodation depend on current law and facts. Current process evidence and qualified HR, employee-relations, accessibility, privacy, security, labor, and legal review control.

Decision test

Ask whether the source changes the decision itself, the evidence required, the implementation sequence, or only the language used to describe an existing capability. Record which claims are directly supported, which are provider statements, which require an independent test, and which remain unknown. A source-linked review should make uncertainty easier to see, not bury it inside a blended score.

Questions to take into review

  • Which policy version and jurisdiction apply?
  • What sensitive topics force escalation?
  • Which roles and tasks change?
  • How are workers or representatives involved?
  • Does the tool materially influence who advances?
  • What validated job criteria support the output?
  • Which data is job-related and known to employees?
  • How are context and accommodations represented?
The publication supports research and executive decision preparation. It does not provide legal, financial, accounting, employment, clinical, cybersecurity, investment, procurement, or implementation advice.