Roundtable Recap: What TA Leaders Discussed Offstage at Moment 2026

Eight candid conversations on fraud, trust, candidate experience, scheduling, change management, connected data, AI skills, and the growing ability of TA teams to build their own solutions.

---

The main-stage sessions at Moment 2026 established the big questions facing talent acquisition. The following morning, attendees broke into eight smaller roundtables to work through the operational realities: fraud, trust, candidate experience, scheduling, change management, connected data, AI skills, and the growing ability of TA teams to build their own solutions.

The conversations were candid, specific, and grounded in what leaders are seeing inside their organizations right now. Together, they revealed an industry moving beyond abstract excitement about AI and confronting the harder questions of how to use it responsibly, effectively, and humanely.

Candidate fraud has become an enterprise risk

Key takeaway: Candidate fraud is no longer an isolated recruiting problem, and there is no single tool capable of solving it.

The examples shared during this discussion made clear how quickly candidate fraud is evolving. Leaders reported fraudulent applications arriving at enormous scale, fabricated resumes and references, AI-assisted interview answers, deepfake interviews, identity substitution, and candidates who performed convincingly online but could not demonstrate the same skills in person.

Participants drew an important distinction between “capital-F Fraud” – organized operations, identity theft, insider threats, and deepfakes – and “lowercase-f fraud,” such as embellished experience, AI-generated answers, and misrepresented skills. Both create real costs, but they require different levels of intervention.

The group also agreed that ownership remains dangerously fragmented. Talent acquisition, legal, security, and executive leadership each see part of the problem, but few organizations have established clear accountability across the full hiring process. TA teams are often handed a new detection tool – or told what they cannot do – without the cross-functional support required to redesign the workflow around it.

• Blocking suspicious IP addresses, email domains, and application patterns
• Adding thoughtful friction at key stages of the application process
• Training recruiters and hiring managers to recognize behavioral changes
• Using identity verification and reference checks as additional signals
• Requiring an in-person interaction or ID check for sensitive positions
• Tracking suspected fraud through legally approved disposition categories
• Escalating the issue as an enterprise risk rather than a recruiting inconvenience

The emerging answer is a layered model similar to cybersecurity. No individual signal is definitive, and no vendor will stay ahead of every new tactic. Organizations need a combination of technology, human observation, clear policies, training, and cross-functional escalation paths.

They must also be careful not to punish every legitimate applicant in an effort to stop a smaller number of bad actors. The challenge is to introduce “smart friction” where risk is highest while preserving a fair and accessible experience for everyone else.

Trust must be earned throughout the hiring process

Key takeaway: Candidates do not automatically distrust AI. They distrust processes that feel opaque, impersonal, or unaccountable.

Trust can begin eroding before a candidate ever speaks with a recruiter. An open role may receive hundreds or thousands of applications, making individualized responses unrealistic. But from the candidate’s perspective, an immediate automated rejection can reinforce the belief that no person ever considered their experience.

The roundtable made an important distinction between applicants and engaged candidates. Companies may not be able to provide detailed feedback to everyone who applies. Once the organization invites someone into an interview process, however, its responsibility changes. Long delays, generic communication, and unexplained decisions can damage candidate trust – and, in some organizations, are already showing up in candidate NPS results.

Legal concerns frequently make the problem harder. Recruiters may want to give useful feedback but are instructed to remain intentionally vague. The group challenged leaders to examine the actual level of risk involved rather than treating maximum risk avoidance as the only acceptable goal. Legal should advise on risk, but executive leadership ultimately needs to decide what balance the organization is willing to strike between protection, transparency, and candidate experience.

AI adds another layer. Participants discussed experimenting with AI screeners that give more applicants an opportunity to explain their experience and answer follow-up questions. Early results have surfaced a clear tradeoff: the technology can create a richer signal than keyword filtering, but candidates may resist if they do not understand why it is being used or what role it plays in the decision.

• When they are interacting with AI
• Why the technology is part of the process
• What information it is evaluating
• Whether a person will review the outcome
• What they can expect at the next stage

The group ultimately agreed that hiring managers should own the final hiring decision – and be accountable for its results. TA’s role is to improve that decision through market knowledge, candidate insight, consistent assessment, process discipline, and the willingness to challenge a pattern that is not working.

AI should create more capacity for human care

Key takeaway: The value of AI is not simply the time it saves. It is what recruiting teams choose to do with that time.

This discussion introduced a useful framework for designing the hiring experience: Basics, Anticipate, Delight.

The basics are the elements every candidate should be able to rely on: clear job descriptions, dependable scheduling, thoughtful communication, accurate travel information, and a process that works as promised. Without these fundamentals, more creative experience initiatives cannot compensate.

The next level is anticipation – recognizing what a candidate or interviewer will need before they ask. Participants shared examples including:

• Preparing candidates for an interviewer’s communication style
• Explaining presentation expectations instead of assuming candidates know them
• Sending executives interview briefs that identify gaps in earlier conversations
• Providing loaner laptops for on-site presentations
• Giving candidates one portal for schedules, preparation materials, and videos
• Creating localized guidance for candidates unfamiliar with methods such as STAR
• Training hiring managers to probe consistently and conduct better interviews

Finally, delight comes from small moments that make people feel expected and valued. Leaders described welcome bags waiting on-site, offer calls attended by the entire interview panel, scorecard excerpts included with an offer, and personalized videos explaining the recruiter’s commitment to the candidate.

The most memorable examples were not necessarily expensive. What mattered was specificity. A generic gift may be pleasant; a gesture that demonstrates someone noticed and remembered you creates a lasting impression.

That principle applies to rejection as well. Thoughtful acknowledgment of the time and energy a candidate invested can change how a difficult decision feels and help preserve trust even when the answer is no.

AI can support this kind of care. Leaders shared examples of using connected interview transcripts and scorecards to generate more personalized feedback, creating stronger executive briefs, and completing market research that previously consumed hours. But the group returned repeatedly to one question: will that reclaimed capacity be used to give recruiters more requisitions, or to create a more human experience?

The closing challenge was intentionally simple: choose one improvement and show up 10% better than before.

Interview scheduling is part of the company’s competitive position

Key takeaway: Scheduling reveals whether an organization is coordinated, responsive, and genuinely committed to hiring.

Interview scheduling is often treated as administrative work occurring between more important stages of the process. The roundtable challenged that assumption.

Delays, repeated reschedules, unavailable interviewers, and unclear communication all shape how candidates perceive the organization. Conversely, a process that moves quickly, respects the candidate’s time, and anticipates their questions can become a real competitive advantage.

The conversation surfaced two fundamentally different approaches to interviewer accountability. Some organizations use firm consequences: one leader described a policy under which a requisition may be removed if a team fails to make progress within 45 days. Others use positive reinforcement through “interviewer citizenship” programs that recognize people who consistently participate and maintain strong calendar practices.

The correct approach depends on company culture. The important thing is that leadership establishes and enforces an approach. An email announcing a new expectation will not change behavior if interviewers see that nothing happens when it is ignored.

• Establish fixed panels or recurring interview blocks for high-volume roles
• Connect interview participation to manager and employee expectations
• Include coordinators in intake meetings before scheduling begins
• Standardize interview plans where repetition makes sense
• Track declines, reschedules, no-shows, and the reasons behind them
• Give leaders visibility into patterns across teams and individuals
• Maintain a clear path for humans to intervene in unusual situations

Standardization, however, has limits. Global interviews, hybrid experiences, executive schedules, accessibility needs, and unexpected candidate circumstances will always produce edge cases. Automation can handle more of the repeatable work, but organizations still need human judgment for the exceptions.

That does not mean the coordinator’s role remains unchanged. As automation handles more calendar work, coordinators can increasingly manage the system around scheduling: automation, data quality, interviewer capacity, recurring bottlenecks, and high-touch candidate moments.

The ultimate goal is not simply a dashboard that lets someone search for problems. It is an intelligent system that notices patterns first – then alerts the team that a region has a decline problem, a new interviewer may need additional support, or a hiring process is beginning to stall.

TA teams are becoming builders

Key takeaway: AI is dramatically lowering the barrier to building new recruiting tools, but sustainable innovation still requires strategy, governance, and adoption.

One of the most ambitious examples came from Remote, which has been building its own internal ATS. The project began with a three-day, hackathon-style offsite and grew into a cross-functional effort involving engineering, IT, legal, compliance, and the people team. With approximately 45,000 applications arriving each month, its requirements were shaped by real enterprise scale.

The discussion made clear that “buy versus build” is not a philosophical choice with one correct answer. Leaders need to consider:

• How quickly the problem must be solved
• Whether existing products can meet the requirement
• The true cost of internal engineering and maintenance
• How much customization creates meaningful value
• Whether the organization has the resources to support the product over time
• What strategic advantage the organization expects to gain

More broadly, participants predicted a “headless” future for TA technology. The ATS remains the system of record, but employees interact with agents and skills through the tools they already use – Slack, Teams, email, and mobile experiences. Hiring managers should not have to learn another interface simply because TA purchased another system.

This creates significant opportunity for TA professionals. Teams are already building internal applications, automations, reporting tools, and agent-driven workflows. But uncontrolled citizen development can recreate the fragmentation AI was supposed to solve. If every team builds independently, organizations are left with undocumented applications, duplicated work, security risk, and tools no one can maintain after the original builder leaves.

• Provide approved frameworks, components, and design systems
• Require security and vulnerability reviews before launch
• Document ownership, dependencies, and maintenance plans
• Create safe spaces for experimentation
• Celebrate small efficiency wins and share them across the team
• Redirect overlapping projects instead of discouraging builders
• Measure whether employees actually adopt what gets built

That last point is critical. Building something impressive is not the same as creating value. The goal is not to turn every recruiter into an engineer. It is to develop enough AI fluency across the team to identify problems, test ideas, collaborate with technical partners, and decide when building something new is truly justified.

AI change management has to reach individual contributors

Key takeaway: Leaders cannot demand AI adoption without helping employees understand what is changing, why it matters, and how they can succeed in the new model.

The most immediate concern employees have about AI is job security. Close behind it are fears that they will not learn quickly enough, that candidates will reject AI-enabled experiences, or that colleagues will judge them for relying on the technology.

That creates a contradictory environment: employees are pushed to use AI but may still be viewed skeptically when they do.

The group agreed that transparency matters, but transparency without careful framing can create more anxiety. Leaders need to speak differently to executives, managers, recruiters, and coordinators because each group has different information, incentives, and concerns.

Responsible transparency pairs an honest description of how work is changing with a credible development path. For recruiting coordinators, that could mean showing how calendar-management experience can evolve into automation ownership, project management, operations, analytics, or program leadership. For recruiters, it may mean developing deeper business knowledge, judgment, relationship building, storytelling, and talent advisory skills.

• Recruit respected skeptics into an early change-advocacy network
• Give employees access to approved tools and safe experimentation spaces
• Run hands-on builder workshops rather than passive training alone
• Establish proficiency levels with clear expectations
• Pair anxious employees with experienced builders
• Create hackathons with dedicated time away from normal responsibilities
• Share and celebrate useful applications across the team
• Make experimentation part of capacity planning instead of invisible extra work

The skills that become more valuable are distinctly human but not vague: judgment, emotional intelligence, business acumen, relationship building, candidate care, organization, coaching, and the ability to translate data into a compelling story.

New hybrid roles are also emerging. Participants described talent enablement engineers, internal AI consultants, agent managers, and recruiters who build tools and workflows for the rest of the team. Not everyone needs to become a builder, but organizations need a clear home for employees who develop that strength.

Measurement should also evolve in stages. Early in the transformation, the most meaningful signals may be participation, training, experimentation, and proficiency. Productivity and business outcomes matter, but forcing an immediate ROI calculation can discourage the learning required to reach them.

Connected data can turn reporting into action

Key takeaway: MCP makes recruiting data more accessible, but easier access does not automatically make the data accurate, consistent, or useful.

The group began with a familiar problem: TA data is scattered across ATS platforms, scheduling systems, HRIS tools, data warehouses, and spreadsheets. Leaders spend hours building weekly reports, estimating recruiter capacity, reconciling conflicting numbers, and trying to answer questions that should be simple.

Even widely discussed metrics can collapse under closer inspection. When one leadership team pushed for a single quality-of-hire score, stakeholders struggled to agree on the underlying definition. Performance ratings arrive late, training scores are imperfect proxies, and different departments interpret success differently.

Rather than forcing a composite metric, participants recommended bringing stakeholders together to list the available signals and assess how much confidence they have in each one. Retention, performance-based terminations, hiring-manager feedback, and early training outcomes may be more useful as a group of adjacent indicators than as one artificially precise score.

MCP can make connecting and querying these systems significantly easier. One dashboard discussed during the session combined GoodTime and Greenhouse data to show coordinator workload, track referrals, detect stalled candidates, and link users directly to candidate records. A usable version was shareable within roughly 20 hours of work.

The advice for getting started was deliberately modest: solve one personal, visible problem first. Build a single chart that eliminates a recurring manual task. Validate it. Then expand.

Accuracy remains the most important guardrail. Teams should require the system to show its work, state its assumptions, and let users drill from summary metrics to candidate-level details. Definitions such as “hire,” “start,” and “time to fill” must be established before the first prompt is written. API access should be tightly scoped, credentials securely managed, and every critical process should retain a human fallback.

The promise is not another static dashboard. It is a connected intelligence layer that can identify a problem, explain the underlying evidence, and help a TA leader determine what to do next.

Hiring for AI skills starts with defining what proficiency means

Key takeaway: “AI proficiency” is too broad to function as a useful hiring requirement. Organizations must define what it means for each role.

Most participants rated their organizations as actively hiring for AI skills but unsure whether they were doing it well. The challenge begins with definition.

AI proficiency may mean basic literacy for one role, using AI to advance domain-specific work for another, integrating models into a product for an engineer, or conducting original AI research at the highest technical level. Applying the same requirement across all of them produces inconsistent evaluation.

One framework discussed during the session separates AI capability into three broad tracks:

1. General AI literacy for nontechnical roles
2. Building with or integrating AI into products
3. Advanced research and orchestration of AI systems

The strongest assessment methods combine behavioral questions with live application. Candidates might be asked to share their screen and demonstrate how they would use their preferred AI tool to approach a real problem. Interviewers can then evaluate how candidates frame the task, test the output, identify errors, and decide where AI is inappropriate.

A particularly revealing area to probe is when a candidate chooses not to use AI. A thoughtful answer can demonstrate judgment and an understanding of the technology’s limits, rather than enthusiasm alone.

Curiosity and adaptability may ultimately be more durable signals than expertise with a specific product. One leader described hiring a recruiter who initially scored zero on an AI assessment but demonstrated a strong desire to learn. After focused training, she became one of the most advanced AI users on a 32-person team within four months.

The conversation also surfaced an unresolved tension around entry-level talent. AI can now perform much of the foundational work through which earlier generations developed expertise. Organizations cannot eliminate those opportunities without also weakening their future talent pipeline. Apprenticeships, graduate cohorts, and redesigned entry-level roles may become increasingly important.

Before the next hiring cycle, leaders should be able to answer three questions:

• What does AI proficiency mean for this specific role?
• What AI use is permitted during the hiring process?
• How will interviewers distinguish genuine judgment from a polished AI-assisted performance?

Without those answers, candidates cannot know what “good” looks like – and hiring managers will continue applying different standards to the same behavior.

The themes connecting the roundtables

Across eight different discussions, several ideas appeared repeatedly:

1.  AI adoption is an operating-model challenge, not simply a technology purchase.

2.  Human judgment becomes more valuable as administrative work is automated.

3.  Trust requires transparency, accountability, and a clearly defined human role.

4.  Connected data is useful only when definitions and underlying inputs can be trusted.

5.  AI creates new opportunities for TA professionals – but employees need a path to develop into them.

6.  Experimentation needs guardrails, ownership, and evidence of adoption.

7.  Candidate experience remains a strategic differentiator, especially as hiring becomes more automated.

8.  The organizations that learn fastest will not necessarily be those with the most tools. They will be those that create the strongest systems for learning, sharing, and adapting.

The conversation is just getting started

If there was one message running through all eight roundtables, it was that talent acquisition is moving from experimentation to execution. The questions are getting more practical, the tradeoffs more visible, and the expectations for TA leaders much higher.

These discussions only captured part of what unfolded at Moment 2026. In the weeks ahead, we’ll be sharing more from the main-stage sessions, including deeper takeaways on enterprise AI adoption, the future of recruiting work, and how leading TA teams are putting these ideas into practice.

For now, the clearest takeaway is this: the future of TA will not be shaped by AI alone. It will be shaped by the leaders who decide how to use it.