
'Please Stop Inviting AI Notetakers To Meetings' 47
Most virtual meeting platforms these days include AI-powered notetaking tools or bots that join meetings as guests, transcribe discussions, and/or summarize key points. "The tech companies behind them might frame it as a step forward in efficiency, but the technology raises troubling questions around etiquette and privacy and risks undercutting the very communication it's meant to improve (paywalled; alternative source)," writes Chris Stokel-Walker in a Weekend Essay for Bloomberg. From the article: [...] The push to document every workplace interaction and utterance is not new. Having a paper trail has long been seen as a useful thing, and a record of decisions and action points is arguably what makes a meeting meaningful. The difference now is the inclusion of new technology that lacks the nuance and depth of understanding inherent to human interaction in a meeting room. In some ways, the prior generation of communication tools, such as instant messaging service Slack, created its own set of problems. Messaging that previously passed in private via email became much more transparent, creating a minefield where one wrong word or badly chosen emoji can explode into a dispute between colleagues. There is a similar risk with notetaking tools. Each utterance documented and analyzed by AI includes the potential for missteps and misunderstandings.
Anyone thinking of bringing an AI notetaker to a meeting must consider how other attendees will respond, says Andrew Brodsky, assistant professor of management at the McCombs School of Business, part of the University of Texas at Austin. Colleagues might think you want to better focus on what is said without missing out on a definitive record of the discussion. Or they might think, "You can't be bothered to take notes yourself or remember what was being talked about," he says. For the companies that sell these AI interlopers, the upside is clear. They recognize we're easily nudged into different behaviors and can quickly become reliant on tools that we survived without for years. [...] There's another benefit for tech companies getting us hooked on AI notetakers: Training data for AI systems is increasingly hard to come by. Research group Epoch AI forecasts there will be a drought of usable text possibly by next year. And with publishers unleashing lawsuits against AI companies for hoovering up their content, the tech firms are on the hunt for other sources of data. Notes from millions of meetings around the world could be an ideal option.
For those of us who are the source of such data, however, the situation is more nuanced. The key question is whether AI notetakers make office meetings more useless than so many already are. There's an argument that meetings are an important excuse for workers to come together and talk as human beings. All that small talk is where good ideas often germinate -- that's ostensibly why so many companies are demanding staff return to the office. But if workers trade in-person engagement for AI readbacks, and colleagues curb their words and ideas for fear of being exposed by bots, what's left? If the humans step back, all that remains is a series of data points and more AI slop polluting our lives.
Anyone thinking of bringing an AI notetaker to a meeting must consider how other attendees will respond, says Andrew Brodsky, assistant professor of management at the McCombs School of Business, part of the University of Texas at Austin. Colleagues might think you want to better focus on what is said without missing out on a definitive record of the discussion. Or they might think, "You can't be bothered to take notes yourself or remember what was being talked about," he says. For the companies that sell these AI interlopers, the upside is clear. They recognize we're easily nudged into different behaviors and can quickly become reliant on tools that we survived without for years. [...] There's another benefit for tech companies getting us hooked on AI notetakers: Training data for AI systems is increasingly hard to come by. Research group Epoch AI forecasts there will be a drought of usable text possibly by next year. And with publishers unleashing lawsuits against AI companies for hoovering up their content, the tech firms are on the hunt for other sources of data. Notes from millions of meetings around the world could be an ideal option.
For those of us who are the source of such data, however, the situation is more nuanced. The key question is whether AI notetakers make office meetings more useless than so many already are. There's an argument that meetings are an important excuse for workers to come together and talk as human beings. All that small talk is where good ideas often germinate -- that's ostensibly why so many companies are demanding staff return to the office. But if workers trade in-person engagement for AI readbacks, and colleagues curb their words and ideas for fear of being exposed by bots, what's left? If the humans step back, all that remains is a series of data points and more AI slop polluting our lives.
Nothing new (Score:2)
This is nothing new. Most meeting software has had auto closed captioning for years. People never looked for it before so they probably didn't realize it was there.
Re: Nothing new (Score:2, Insightful)
Re: Nothing new (Score:5, Insightful)
Yeah, if you don't care at all about completeness and accuracy, it's great.
Re: (Score:3)
Yeah, if you don't care at all about completeness and accuracy, it's great.
Completeness and accuracy are in the eye of the beholder.
I learned many years ago that there's great power in volunteering to take notes. He who takes the notes controls that narrative.
Re: (Score:2)
Yeah, if you don't care at all about completeness and accuracy, it's great.
Completeness and accuracy are in the eye of the beholder.
I learned many years ago that there's great power in volunteering to take notes. He who takes the notes controls that narrative.
You are very correct, and wise beyond your years.
Re: Nothing new (Score:5, Interesting)
They are god send for obscure managers inclined to misinterpret everything said.
AIs aren't good at sarcasm or irony.
I'm just waiting for the first person fired due to some bad AI recap.
Re: Nothing new (Score:2)
Re: (Score:2)
Auto closed captioning and live translation is something that can be quite useful in certain fields, but meeting software vendors such as Microsoft have once again shot themselves in the foot when it comes to marketing and selling these features. At work we have some sort of Microsoft Super Premium Plus whatever subscription that includes Teams Premium Pro Plus something or other... but when you try to activate live translation in a call with a Chinese supplier, it turns out that you need a Teams Premium Pr
or..... (Score:5, Insightful)
Re: (Score:2)
If they put their words into an email or a memo, it might be leaked! The powerful, world-shattering revelations revealed by these meeting savants is perhaps too powerful for the uninitiated to behold. AI, still in its infancy, is clearly incapable of capturing the full majesty on display in the body and voice of gods.
Re: (Score:2)
People don't read emails anymore either. These things in combination point to potentially larger issues for an organization.
AI in my job interviews (Score:5, Interesting)
I'm on the interviewing team at my work. All interviews are via Zoom and now there's an AI notetaker. I had hopes that it would alleviate me my own notes to fill in on the overly busy assessment page. It vomits content all over the assessment to the point of being useless. By way of comparison CoPilot vomits lightly when I'm coding and can get some value tidying up the suggestions. So far, the AI for meetings and interviews is only good at light summaries, rather than deep distillation into very subjective human categories. Interviews will start to get gamed with these tools, just like Leetcode made interviewing lazy.
Re: (Score:2)
Meetings are more about inclusivity and management by committee than anything else these days. If the final-say person ever joined a meeting it would be different...but most meetings are all the people debating how to spank the monkey and then the boss gets 5 minutes with 2 people that have their ear and decides wither what they tell 'em or some random batshit idea they already had.
TL;DR you can't spank the monkey by committee, the monkey will ultimately spank the committee.
whining? (Score:2)
I believe I am willing to work in these conditions so that government and finance organizations will be pressured to do the same. I don't trust 90% of the people who are in decision making positions in those organizations and feel they should have to wear microphones with recording at all times... Even in the shower.
Inviting spies to your meetings (Score:5, Insightful)
These AI notetakers aren't "free" for nothing. By inviting them to your meeting, they get access to the names and contact information of every person attending the meeting, and also to everything that is said during the meeting. It's like inviting a spy from an unknown company with unknown intentions, to eavesdrop on everything that happens. Not smart!
Re: (Score:2)
So I clicked on his history.
He seems very normal, unfortunately.
Whereas you are posting weird claims as AC. Maybe it's you who have something to hide?
Disney (Score:3)
Oddly, certain local government employees found out that playing Disney music during an interaction with the general public prevented the video recording from being shared on social media.
Will Disney also go after meeting recordings where anyone playing a music or video clip from one of the Disney movies?
There's the problem of every single word being retrievable...
https://www.cnn.com/2023/08/08... [cnn.com]
Regulators fine Wall Street firms $549 million for using WhatsApp and other channels to discuss business
By Matt Egan, CNN - Published 11:06 AM EDT, Tue August 8, 2023 - New York CNN —
Wells Fargo and a slew of other Wall Street firms admitted Tuesday to using WhatsApp, Signal and other messaging platforms for “off-channel” communications in violation of federal recordkeeping requirements.
The Securities and Exchange Commission said the Wall Street firms acknowledged wrongdoing and have agreed to pay penalties totaling $289 million.
The Capitol Dome and the West Front of the House of Representatives are seen in Washington, Monday, April 17, 2023. (AP Photo/J. Scott Applewhite)
The SEC said its investigation uncovered “pervasive and longstanding ‘off-channel’ communications” at Wells Fargo, BNP Paribas, SG Americas, BMO Capital Markets, Mizuho Securities, Houlihan Lokey, Moelis, Wedbush and SMBC Nikko Securities America.
According to regulators, those firms admitted that from at least 2019, their employees often communicated about business through WhatsApp, iMessage, Signal and other messaging platforms on their personal devices. The SEC said the firms violated federal securities laws by failing to maintain or preserve the “substantial majority” of these communications.
Re: (Score:2)
Oddly, certain local government employees found out that playing Disney music during an interaction with the general public prevented the video recording from being shared on social media.
Those employees probably just found a way to make it legal for you to air Disney music, because broadcasting the official record for news or political reasons would be fair use. Absolutely 1st-Amendment protected speech to retransmit that, and the government agency employing those people would be legally responsible fo
Re: (Score:2)
"Cynicism is the only form in which base souls approach honesty." Friedrich Wilhelm Nietzsche. Or they could just not delete their emails, that's the legal thing to do.
be wary of companies that don't like documentation (Score:2)
[...] The push to document every workplace interaction and utterance is not new. Having a paper trail has long been seen as a useful thing, and a record of decisions and action points is arguably what makes a meeting meaningful.
If a company has an email deletion policy, it is almost certainly because they are trying to hide things from lawsuits (or worse). Having past records of discussions is just too useful, any sane person would keep them around unless they were worried about hiding something.
Re: (Score:2)
If a company has an email deletion policy, it is almost certainly because they are trying to hide things from lawsuits (or worse).
Funny how you can be right while still being wrong. (also fuck off "help me write" AI prompt) Companies delete emails for two primary reasons:
1 - liability. Hindsight isn't 20/20 it's (for lawyers, the opposite of) rose colored glasses. Plenty of things that aren't and weren't illegal still can stir the pot during discovery. Someone asks "hey, can we just cheat this because we know the answer is always 42? Fuck no. We need to be responsible and this is your 3rd week on the job. Stop this or be fired
Re: (Score:3)
1 - liability. Hindsight isn't 20/20 it's (for lawyers, the opposite of) rose colored glasses. Plenty of things that aren't and weren't illegal still can stir the pot during discovery. Someone asks "hey, can we just cheat this because we know the answer is always 42? Fuck no. We need to be responsible and this is your 3rd week on the job. Stop this or be fired" and discovery latches onto fraud. Meh. Also if you don't have it, it's not discoverable - oh well, too bad, shit happens.
The funny thing is, you don't realize that what you've typed is exactly the same as "they are trying to hide things from lawsuits."
Re: (Score:2)
You get involved in litigation and it proceeds to discovery. The opposing party makes a request for some class of records. Now IT(or som
Re: (Score:2)
Anyone who reflexively avoids discoverable channels raises significant red flags; as do particularly aggressive retention policies(yes, 'retention policy' is almost always more about deletion; but they are still always called that);
Yeah, I stopped calling them "retention policies" and started calling them "deletion policies."
#triggered (Score:2)
Sorry, but I can no longer participate in meetings with active AI participants such as notetakers. It's against my deeply held religious beliefs that people are too fucking lazy and if you can't attend a meeting but it can happen with out you then you aren't important. Oh, i mean it's against christ our lord and saviors and detailed in matthews 22:50:3a subsection b12
don't overlook accessibility (Score:1)
just do it (Score:4, Interesting)
Don't let AI do this. Sure you will have meeting notes, but these are useless. Dig in. Do your due diligence. It is slow, it is expensive, some managers will discourage it. But it will pay off. Before you know it, people will start to really cooperate as a team.
Re: just do it (Score:1)
Re: just do it (Score:2)
Extremely tight deadlines. Agitated discussions when things got delayed. Promising miracles to cu
Re: (Score:1)
I always make notes during meetings. I always refine them afterwards. It is a good excuse to reflect on what is said and to get new ideas. Sending them around to ask for feedback or corrections starts the same reflection process with others. Summarizing them at the start of the next meeting gets people to focus on the important stuff. Don't let AI do this. Sure you will have meeting notes, but these are useless. Dig in. Do your due diligence. It is slow, it is expensive, some managers will discourage it. But it will pay off. Before you know it, people will start to really cooperate as a team.
(shrug) One of our clients uses an AI meeting summarizer. We were dubious, but didn't think it was worth offending them over.
In practice, it's actually not bad at summarizing the meetings. Much better than we expected.
Re: (Score:2)
There are the ones (normally fairly expensive) that are sold by the makers of the meeting software and run directly in the security context of whatever participant is using them; but a lot o
All these AI tool from OpenAI, Microsloth, Google (Score:3)
All these AI tool from OpenAI, Microsloth, Google and friends are nothing more than annoying clippys from the last century. The only improvements I see in the past few decades is voice recognition and text to speech has improved quite a lot. These AI tools can help frame and refine sentences and can help come up with nice sounding paragraphs. Image generation capability from descriptions is also very good. These assistants are more convenient and time saving than a search engine, provided you do the fact checking.
Beyond that they are mostly useless. I have used the latest and greatest coding assistants at my work and they all suck.
Applied stupidity (Score:2)
The key task in taking notes in a meeting is to distill out the important points. That takes insight and understanding of context. LLMs cannot do context they know nothing about and struggle with context they know (i.e. they usually mess it up), and no AI can do insight at this time (and maybe forever).
All these "tools" accomplish is taking worthless notes that waste time later on.
The logical conclusion to AI for meetings? (Score:2)
As soon as these things started showing up in meetings it made me think of this scene from Real Genius (80s movie).
https://www.youtube.com/watch?... [youtube.com]
Is this what AI is leading us to?
Worse, the AI produces managerial metrics (Score:3)
I gave a tutorial, an AI notetaking/summarizing tool then produced an "Engagement" metric, with a low 30%. It did not say how this was produced, only hints in their FAQ. Basis included # of cameras turned on, % of eyes watching screen versus off-screen, and # people talking.
There were 50+ people but for the tutorial only 3 had cameras on: me, the moderator (frequently looking off-screen) and a Guy. So if the Guy looked away, that plunged the "Engagement" metric. Plus that it was a Tutorial (95% me speaking) and *poof* low engagement.
My fear is management will latch onto these undocumented AI-generated Metrics and use them in performance evaluations. "Get your scores up, A., 30% is terrible!" But it's not a real measure!
Re: (Score:2)
https://www.nytimes.com/2023/1... [nytimes.com]
Useful for placing blame (Score:2)
When things go wrong, bosses can troll through these notes to find someone to blame.
Otherwise useless.
retarded advice (Score:1)