The News
Linda Galindo, an accountability consultant and creator of the trademarked brand The Voice of Accountability®, is pitching a pointed critique of AI-generated workplace communications, specifically performance reviews, warnings, and termination notices. Her argument is grounded in a pair of findings that at least 26% of employees suspect they’ve received an AI-written performance review, and that 19% of employees who suspected an AI-written warning or layoff email reported crying after receiving it. Galindo’s position is that deploying AI to handle inherently human accountability conversations is not an efficiency gain but a failure of leadership dressed up as innovation.
Analyst Take
This pitch arrived in an HR tech editor’s inbox, but the real audience is every IT and engineering leader who has started routing sensitive people-management tasks through a large language model because the interface is fast and the output sounds professional. That convenience is exactly the problem Galindo is naming.
The AI-in-HR Question Is Actually an Accountability Governance Question
The 26% suspicion rate cited in this pitch is a signal worth pausing on. When more than one in four employees believes their performance review was written by a machine, the document’s credibility collapses regardless of its accuracy. Performance feedback works because it implies that a specific person observed specific behavior, formed a judgment, and chose to communicate it. Remove the human from that chain and you haven’t streamlined a process. You’ve gutted the process of its only operative ingredient.
Galindo’s framing maps cleanly onto a governance gap that engineering and IT organizations are currently navigating. Deploying AI in HR workflows is a subset of the broader AI governance challenge, and ECI Research’s 2026 Application Development survey data shows the industry knows it with 58.2% of respondents saying they will increase AI governance spending by a moderate 10–25% in the coming period. That’s a meaningful commitment. But governance investment focused on model risk, data provenance, and compliance will not automatically cover the softer but equally consequential question of where AI should not speak on behalf of a person.
What Developers and Platform Teams Are Actually Building Into
The scenario Galindo describes is not a hypothetical edge case. AI-assisted writing tools are now standardized across teams in many organizations. According to ECI Research’s 2026 Application Development: Day 0 survey, 52.6% of respondents said AI-assisted coding tools are used in a standardized way across teams. That number almost certainly understates the use of AI writing tools in adjacent workflows, including HR communications drafted on the same platforms developers already use daily. The pipeline from “help me write this code comment” to “help me write this performance warning” is shorter than most IT leaders want to admit.
Platform engineers and ITDMs building internal developer platforms and employee-facing tooling need to think carefully about where they draw the boundary between AI augmentation and AI substitution. Augmentation means a manager uses AI to organize their thoughts, check tone, or surface relevant policies. Substitution means the manager clicks send on output they did not personally author, about an individual they are accountable for managing. Galindo’s argument is that the second pattern is not a productivity gain but an abdication. The 19% of employees who cried after receiving what they suspected was an AI-generated termination notice are the real cost that doesn’t appear in any efficiency metric.
The Accountability Architecture Problem
Galindo’s three-part framework for real accountability (a specific person named, skin in the game, consequences stated upfront) is not a soft leadership platitude. It’s a design requirement. Any system that replaces those three elements with generated prose has not automated accountability. It has automated the appearance of accountability while distributing ownership to no one in particular. For organizations serious about AI governance, this deserves explicit policy treatment; not just “AI must be reviewed before sending” but “the named manager must be able to defend every sentence in this document as their own judgment.”
Looking Ahead
The HR tech market is moving fast toward AI-assisted performance management, and vendors will keep shipping features that make it easier to generate feedback at scale. The backlash from employees will be proportional, and organizations that treat that backlash as a communications problem rather than a design problem will keep getting it wrong. Galindo’s critique will find an increasingly receptive audience as the gap between AI-generated polish and genuine managerial accountability becomes more visible to the people on the receiving end.
For IT and HR technology buyers, the strategic question over the next 12 to 18 months is not whether to use AI in people management workflows, but where to draw an explicit line between assistance and authorship. Organizations that define that line in policy, enforce it in their platforms, and train managers to own the output rather than outsource it will be in a fundamentally stronger position than those that conflate speed with accountability. The ones that don’t will keep generating efficiency metrics while quietly eroding the trust that makes their organizations function.
Stay Ahead of Application Development Trends
Get weekly analyst insights, research notes, event coverage, and AppDevANGLE updates delivered directly to your inbox.
Subscribe for Weekly Insights
Join technology leaders, practitioners, and GTM teams following the trends shaping modern software delivery.
Looking for deeper research access?
Explore ECI Research reports, survey insights, and market analysis through the ECI Research Portal.
