Has AI Ruined Meetings?

By Jennifer Cho

Has AI Ruined Meetings: A Leader’s Guide to What’s Really at Stake

Somewhere along the way, our meetings became crowded with invisible guests. 

Zoom AI Companion, Fathom, Read.ai, Otter.ai, and Fireflies.ai are just some of the AI meeting data tools that now join video calls uninvited, cheerfully announcing they’re “recording and transcribing for your convenience.” And suddenly, nothing feels quite off the record anymore.

I can’t help but wonder: Has AI ruined meetings? I suppose that depends on how we define “ruined.”

AI has fundamentally changed meeting power dynamics, the legal landscape, and the expectations of everyone in the room. And pretending otherwise is how companies drift into real risk.

This isn’t a panic piece. It’s a clear-eyed look at what’s actually happening, and what CEOs, CFOs, and boards need to get in front of before the consequences arrive uninvited.

The Upside: AI Makes Meetings More Productive… and More Searchable

Let’s give credit where it’s due.

AI-generated meeting summaries and action items are fast, clean, and helpful, especially for leaders juggling teams across time zones, calendars, and relentlessly shifting priorities.

Tools like Zoom AI Companion, Fathom, and Otter.ai do a good job of:

• Capturing decisions and commitments (usually) without error

• Identifying action items and surfacing them automatically

• Providing transcripts for those who couldn’t attend

• Improving presentation skills with reports on how many filler words were used, or who (over)dominated the conversation

These tools can aid in sales and customer service trainings, and serve as a resource to marketing and communications leaders. 

For CFOs and finance teams, the appeal is obvious: fewer missed details. Fewer “I thought you said…” moments. A clear audit trail of decisions.

If meetings used to disappear into the ether, AI brings order, structure, and clarity. But that clarity comes with a cost: the data these tools produce is now a permanent, searchable record, owned by someone, stored somewhere, and subject to rules most companies never review.

Who Owns AI Meeting Data?

AI meeting data ownership is where the conversation gets less convenient.

When an AI bot joins a meeting, the transcript doesn’t necessarily belong to the company that hosted the call. It might belong to the tool provider. Or it might be co-owned. Or it might be stored indefinitely on a server outside your state, or outside the country. 

It’s up to consumers to carefully read the terms of service, and even then, it gets murky when “some platforms reserve the right to use aggregated anonymized data for service improvements,” opening the door for potential misuse.

In summary, if you don’t know who owns your AI meeting data, you’re not alone. Most companies don’t.

If your business is in California, you also have to consider the California Consumer Privacy Act (CCPA), which gives meeting participants certain rights over data collection, storage, deletion, and access. And that’s before we start talking about intellectual property risk, confidential financial discussions, or the fact that these transcripts can be pulled into discovery during litigation.

The question executives should be asking is not “Is this convenient?” It’s: “Where is this data going, and who controls it?” Because whoever owns the transcript owns the narrative. And whoever stores it determines the security.

The Consent Problem: Are People Really Free to Say No?

Most of us have been in a meeting where someone says, “I’m turning on Fathom to capture notes, everyone okay with that?”

Is anyone actually going to say no? Or click “no” when prompted. Probably not. Power dynamics don’t disappear just because the bot claims it’s a “silent participant.”

This creates a subtle but significant leadership issue:

Can your team speak candidly if everything they say becomes part of a searchable, permanent record?

That’s the real cost. Meetings designed for open dialogue become carefully scripted exchanges no one wants quoted back to them later.

AI isn’t stopping people from talking. It’s stopping people from saying what they actually mean.

The Legal Shadow: What No One Is Thinking About (Yet)

Here’s the part most companies underestimate:

Meeting transcripts are discoverable.

If your company is ever involved in litigation, internal investigations, or regulatory inquiries, those transcripts may be pulled, even ones you forgot (or didn’t even know) existed.

Consider the conversations that happen in an average week: budgeting, strategy, and performance discussions. Now imagine all of that being indexed, time-stamped, and searchable.

Suddenly, what’s convenient today is potentially explosive tomorrow.

And if an AI tool stores data outside your system, your legal exposure may not even be within your control.

This is why CFOs, who already think in terms of risk, compliance, and governance, must be part of AI policy conversations. AI is not just an IT or operations concern. It touches finance, legal, HR, and brand reputation.

The Environmental Cost of AI: The Footprint Nobody Sees

AI feels weightless. Bots join calls, summaries appear, and everything lives in the cloud. But none of this is weightless. 

Behind every AI meeting transcript is:

• A data center consuming enormous amounts of energy.
  Training a single large AI model can produce about 626,000 pounds of carbon dioxide, equivalent to approximately 300 round-trip flights between New York and San Francisco,

• Cooling systems burning through water. A large data center can consume as much water as 6,500 homes do in a day. Data centers globally are expected to use 560 billion liters of water this year, and that figure may double by 2030.

We’re tech optimists in Silicon Valley, but we can’t ignore the math. AI demand is growing faster than the grid can realistically support without real environmental consequences.

Every meeting recorded, transcribed, processed, and stored generates energy costs. At scale, the footprint is enormous.

That doesn’t mean we stop using these tools. It means leaders need to understand that AI convenience comes with invisible infrastructure and environmental trade-offs we rarely acknowledge.

So, Has AI Ruined Meetings? No. But It Has Changed Them.

For leaders, the real problem isn’t the technology. It’s the vacuum of governance around the technology. Meetings aren’t ruined, but they are unregulated. And unregulated systems create avoidable risk.

If CEOs don’t set the standard, the standard sets itself.

What Leaders Should Do Now (Without Getting Overwhelmed)

You don’t need a 90-page AI policy to get started. You just need structure, clarity, and accountability. 

1. Decide when meeting recording tools are permitted. Board meetings? Consider pros and cons. Cross-functional working sessions? Maybe and decide when it’s appropriate. Client calls? Most likely not, but only with explicit consent.  Interviews?  Likely no.
   
   
2. Establish ownership and retention rules

   – Who owns the data?
   – Where is it stored?
   – For how long?
   – Who can delete it?
   – Create a policy and share it company-wide.

3. Involve the CFO in AI governance.

   AI touches compliance, risk exposure, and operational efficiency. Finance leaders have to be at the table, not brought in once the damage is done.
   

4. Train managers on consent and confidentiality
   – “Anyone mind if I record?” is not enough.
   – Explain the implications. Offer alternatives.
   – Make non-recorded meetings normal and standard.

5. Review your vendors
   – Don’t assume a popular tool is compliant.
   – Know the company behind the bot.
   – Know their security standards.
   – Know their deletion policies.
   
6. Keep space for real human conversation

Not everything needs to be transcribed. Leaders should protect the ability to think aloud, disagree, and explore ideas without an AI remembering every word.

Final Thoughts

AI isn’t going away, and neither are the tools quietly joining our weekly calls. They will continue to evolve, get smarter, and slide deeper into the everyday fabric of how we communicate and do business.

AI has already reshaped meetings. It’s now up to leaders to shape how AI is used, or we let the technology make the rules for us.

This is one of those moments where calm, thoughtful governance pays dividends. Set your policies now. Understand your tools. Protect your people’s ability to speak freely. And use AI where it genuinely adds value. without surrendering your judgment to it.