Delivering bad news is one of the hardest parts of managing people. Telling someone their performance is falling short, that the raise they expected isn’t coming, or that their job is being eliminated ranks among the most dreaded tasks in the workplace. Now a growing category of AI coaching tools is aimed at the preparation phase — the hours before a manager sits down with an employee for a conversation nobody wants to have.
The need is real, according to a July survey from The Predictive Index, which offers a text-based chatbot called Obi that helps managers role-play and prepare for difficult conversations. While 74% of CEOs and business leaders said their managers are “very confident” handling tough people conversations on their own, 42% of managers said they know what they want to say but struggle with how to say it.
Only 20% of managers said they prefer to prepare for a difficult conversation alone, without a framework, a coaching guide or input from HR. The survey identified managers’ top blind spots as delivering criticism constructively, anticipating an employee’s reaction, and staying objective.
Preparation, not improvisation
The tools on the market vary widely in features and price. Some are free, others paid; some offer text-only interactions while others support voice. Certain products incorporate employee data such as personality profiles to make the AI’s simulation more realistic, while others stay generic. What they share is a purpose: letting a manager rehearse the message and simulate how a real person might respond, rather than simply running through the script alone.
“Spontaneity is not a goal of a high-stakes work conversation,” said Emily DeJeu, a professor of business communication at Carnegie Mellon University’s Tepper School of Business. “You want to know the executive has really thoughtfully prepared for an important conversation.”
DeJeu said practicing with an AI tool beats talking to a mirror, because the manager gets feedback in return. Managers can also run multiple versions of the same conversation. In one round they might ask the AI to play an employee who reacts angrily; in another, someone who becomes emotional. Karan Kashyap, co-founder and chief executive of Posh, which offers an AI role-play training tool for businesses, said rehearsing several scenarios helps “because you don’t know what real life will throw at you.”
John Morgan, president at the human resources consulting firm LHH, which offers an AI coaching tool through a subsidiary, compared the drills to batting practice: the only way to improve is repetition, and real-world chances are limited. The tools are also available around the clock. “Most companies aren’t hiring 500 coaches for 500 managers,” said Casey Schaffer, founder of Definitions Coaching and Consulting.
Specific instructions get better results
The technology performs best when given parameters and context. A manager could ask the tool to role-play giving feedback to a sensitive employee who is habitually late, for example. Megan Barbier, chief human resources officer at Xactly, whose text- and voice-based tool Penny provides tailored scripts and real-time guidance, said managers can ask to be graded on specific measures, request critical feedback on their message, or have the AI flag language that strays into legally risky territory.
According to Anthony Belluccia, a senior product scientist at The Predictive Index, the role-playing itself is useful, but the real lessons come from the feedback — chatbots can deliver it in a constructive, tailored way so managers see what they could do better.
Without that feedback, many managers don’t realize how their tone or word choice lands. DeJeu described a manager who opened difficult conversations with the bad news rather than easing into it. Before the AI pointed it out, he hadn’t recognized that his blunt delivery might put employees off. The tool advised him to start by discussing the employee’s contributions and value to the company before moving to the negative news. “It guided him on how to reframe the conversation,” DeJeu said.
AI can also flag when managers talk too much, repeat themselves or lean on filler words like “um” and “so” that disrupt the flow. “It’s a gift to other people when you can speak clearly and effectively,” DeJeu said. “It just takes a lot of work to get there.”
Employees can use the same approach. If someone senses a tense conversation coming, preparing in advance with a free tool such as Claude or ChatGPT can lead to more thoughtful and effective responses.
Guardrails and human backup
One risk is that AI tools are built to be agreeable — exactly the wrong instinct when a manager wants constructive criticism. DeJeu advises telling the tool to be more critical and direct if it’s being too accommodating.
Privacy is another concern. HR professionals generally recommend against entering personal details into AI tools. Schaffer said she would “avoid anything that has any identifiable information in it,” suggesting managers describe behaviors or anecdotes instead of names and specifics. Managers should be especially careful about putting employee names, performance details or confidential information into public tools, while employer-approved enterprise systems may carry different privacy controls and authorized uses that managers should understand. Vendor policies on data retention, access and AI model training also deserve scrutiny.
Even with AI in the mix, experts say a human coach remains valuable. Morgan noted that managers sometimes need a perspective only a person can provide, especially in complicated situations. Schaffer recommends starting with an AI coach, then debriefing with a supervisor or HR team member to catch anything missed and to rehearse the message with a live person. “The emotional weight of something like a termination conversation is nearly impossible to replicate,” she said.
Source: www.cnbc.com — https://www.cnbc.com/2026/09/12/ai-manager-coaching-employee-feedback-performanc-review.html
This article is for informational purposes only and does not constitute financial, investment, tax, or legal advice. Do your own research and consult a licensed professional before making financial decisions.
