When AI Has No Guardrails
AI lacks safeguards protecting women and girls
Sep 7, 2026 · 1 hr 25 min · 15 segments
Artificial intelligence may be the most transformative technology of our generation—but according to Fern Potter, AI alone is not a strategy.In this episode of Signal & Noise, hosts Rio Longacre and…
Rio LongacreHostAnd so the one that really annoys me is people using it for their LinkedIn posts.
)
Ference, one task in your life that you either already or would happily delegate to AI.
I mean, one that I would love to delegate is my tax return, but I don't think we're there yet from a government perspective.
But definitely, the thing I do love is just all the meeting recordings, all that qualitative data that you can get in a single place right now and be able to look back on, but also have a singular record on that feeds your own knowledge base.
and there's a lot of cool use cases in AI around that, around not just creating a folder with meeting notes that have been carefully curated by your AI bot, right? Uh, and, and, you know, and then going, but something where it just absorbs all of that across all of your clients and then allows you to, to conversational interface.
We're actually building something like that internally within, within our org.
But, but point being is that, that I just think there's, it's again, it's like, how do you leverage that intelligence of all of your meetings with a client over time, as opposed to having to ask the LLM to go and review all the files all over again.
It sort of builds a knowledge graph around that particular client and all of the interactions you've had over time with them.
And it doesn't waste all the tokens that it would take to just keep doing that analysis every time you want to do an evaluation.
But yeah, and even like, you know, you've got these, what do I use? You talk to yourself, those dictaphones and stuff as you're going around and be able to put that in together with your ideation.
And so the one that really annoys me is people using it for their LinkedIn posts.
)
Ference, one task in your life that you either already or would happily delegate to AI.
I mean, one that I would love to delegate is my tax return, but I don't think we're there yet from a government perspective.
But definitely, the thing I do love is just all the meeting recordings, all that qualitative data that you can get in a single place right now and be able to look back on, but also have a singular record on that feeds your own knowledge base.
and there's a lot of cool use cases in AI around that, around not just creating a folder with meeting notes that have been carefully curated by your AI bot, right? Uh, and, and, you know, and then going, but something where it just absorbs all of that across all of your clients and then allows you to, to conversational interface.
We're actually building something like that internally within, within our org.
But, but point being is that, that I just think there's, it's again, it's like, how do you leverage that intelligence of all of your meetings with a client over time, as opposed to having to ask the LLM to go and review all the files all over again.
It sort of builds a knowledge graph around that particular client and all of the interactions you've had over time with them.
And it doesn't waste all the tokens that it would take to just keep doing that analysis every time you want to do an evaluation.
But yeah, and even like, you know, you've got these, what do I use? You talk to yourself, those dictaphones and stuff as you're going around and be able to put that in together with your ideation.
Every episode on Radar is fully transcribed, speaker-labeled, and rich with metadata. Here is a taste of this one. Try Radar for free to see the rest.
3 of 12
When AI Has No Guardrails
AI lacks safeguards protecting women and girls
AI Needs Human Intelligence
AI needs a human definition of success
Bias Can Widen Existing Gaps
AI can amplify bias without human governance
+9 more clips · 11 min 38 sec of audio in all
7 of 25
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
All 12 clips — the highlight moments, each cut as its own audio, with a title and a speaker
All 15 segments — the transcript broken into labeled sections, every ad read marked
All 25 topics — jump to every other episode discussing the same subject
Every related episode — other shows Radar links to this one
No account is needed to search Radar.