ER doc focused how humans and systems keep working when things go wrong. RAND | USC | MCTI | lead: @TheEmergMind
Joined December 2011
- Tweets 415
- Following 626
- Followers 741
- Likes 402
13 Photos and videos
Pinned Tweet
Emergencies aren’t just worse versions of bad days-qualitatively different cognitive states, team dynamics, failure modes.
You can’t prepare for them by just doing more of what you already do.
You have to train (and train your systems) specifically for crisis.
1
2
7
408
What does building trust with an AI system have to do with supervising a medical intern?
More than you'd think.
Episode 135 of @TheEmergMind podcast with Dr. Sarah Gebauer on trust, failure, and AI governance.
⬇️⬇️
youtu.be/FP0VX32AJX8
1
1
2
222
The intern analogy is real though. You probe the edges of what they know. You ask the critical questions. You give increasing autonomy as trust is earned. That structure translates. What doesn't translate is assuming the failure points will look familiar.
1
1
16
Episode 135: Dr. Sarah Gebauer on trust, failure, and AI governance.
Hope you enjoy!
#AISafety #AIGovernance #SangfroidLabs
youtu.be/FP0VX32AJX8
1
19
Let's talk psychology of swarm teams.
Operating as part of a swarm team is different than working on a normal team. And swarming is a distinct skill that requires thought and practice.
psychologytoday.com/us/blog/…
1
3
4
273
[3] Meaning: When the team dissolves, swarmers are left to process what happened alone.
No formal elders, no debrief structure, no shared community.
This can be lonely and challenging, and requires a proactive response by the groups that swarmers return to.
1
1
20
Systems that swarm need to be attuned to these psychological aspects of swarming before the next crisis starts, not just during it.
Would love to hear what you think.
#SangfroidLabs #EmergencyMind #SwarmTeams
psychologytoday.com/us/blog/…
1
2
36
Dan Dworkis MD PhD FACEP retweeted
May 29
LLMs are affecting what people do in emergencies (and not for the best).
More safety research needs to be done by medical professionals. I want to see @ddworkis working with frontier labs (@OpenAI / @AnthropicAI) to improve the way LLMs are trained in this space.
If you woke up with a numb or weak arm and asked an AI what to do, would it tell you to call 911?
It should. But the answer is: depends which one you asked. And on exactly how you asked it.
medrxiv.org/content/10.64898…
A thread on #AIsafety, #healthcareAI, and #stroke ⬇️
2
5
418
If you woke up with a numb or weak arm and asked an AI what to do, would it tell you to call 911?
It should. But the answer is: depends which one you asked. And on exactly how you asked it.
medrxiv.org/content/10.64898…
A thread on #AIsafety, #healthcareAI, and #stroke ⬇️
1
1
4
694
All three models got distracted by a potential alternative explanation. They also showed strong word-to-word differences, recommending 911 far more often for "weak" than for "numb."
2
1
82
We really don't know how patients actually describe what they're feeling when having a stroke.
We know "weak" and "numb" are scary, but we really don't understand at the verbatim level of granularity we need.
@AbridgeHQ | @AmbienceAI | @corti_ai - big opportunity to help here!
21
Which model they use and what words they happen to choose to describe what they're feeling could send them in completely different directions with irreversible consequences.
Preprint below - curious what you think.
medrxiv.org/content/10.64898…
1
27
Big thank you to co-authors: Richard Lucarelli, @JonStenstrom and @ayan2024
1
1
3
115