Chris Kennedy

{Psychiatric, causal, psychometric, robotic} AI. @MGHPrecisionPsy @MGHPsychiatry MGH-Biostatistics @HarvardMed; @UCBerkeley biostat PhD

[Chris Kennedy @c3K]

Latest suicide pub (w/ @drkatebentley @TaylorABurkePhD @jorsmo @DrJBStephens) provides strong evidence to 1) counter the common claim that clinicians can't predict suicide risk, demonstrating the value of clinical judgment, yet also 2) massive improvement through ML on SRA items

[JAMA Psychiatry @JAMAPsych]

While clinicians can stratify suicide risk above chance levels, predictive accuracy for future suicide attempts significantly improves when using machine learning to incorporate comprehensive clinical assessment data.

https://t.co/QvbBKljIa8 @drkatebentley

[Chris Kennedy @c3K]

it turns out that when a researcher is in hell, they are just forced to setup/link scienv, orcid, and web of science repeatedly 😭

[Chris Kennedy @c3K]

@JeremyNguyenPhD @acagamic Sure, but I might not lead with "OLS, logit, OR probit". You could just say "regression" and move on

[Ramin Nasibov @RaminNasibov]

I don't know why you need this information, but this is the aerodynamics of a beaver

[Chris Kennedy @c3K]

@ylecun @aminkarbasi @CSProfKGD Hmm so a world model + optimal dynamic treatment estimator + agent?

[Chris Kennedy @c3K]

@SashaGusevPosts What is a movement if not a reasonably coherent bundle of issues seeking to break into the fleeting zeitgeist?

[Chris Kennedy @c3K]

Maybe knowledge was never what teaching was about. Maybe it was about inspiring, asking questions, sharing enthusiasm, mentoring, and coordinating group work. AI aside, the flipped classroom educational approach already demonstrated that straight lecturing is not a great strategy

[Steve McGuire @sfmcguire79]

“AI is demoralizing.”

A Princeton Professor says he kept wondering this semester (while lecturing) if his students would be better off learning from Claude:

[Chris Kennedy @c3K]

@JohnHolbein1 I'd be open to contributing to this btw

[Chris Kennedy @c3K]

@JohnHolbein1 I think you mean estimate, not measure. Or do you get into the psychometrics of ideology?

[Chris Kennedy @c3K]

@sahnicboom @NeilOBrian4 and nice to see more translational polisci! congrats Neil

[Chris Kennedy @c3K]

@ForensicPIOrg Agreed, but are you arguing for citation integrity, or just citation integrity for LLM-involved content?

[Chris Kennedy @c3K]

@Manigarm @thomasmetcalf How do you reconcile that argument with a senior author who provides substantive, thorough feedback on a paper?

[Chris Kennedy @c3K]

@ForensicPIOrg @thomasmetcalf But humans also create referencing errors on their own

[Chris Kennedy @c3K]

@DrSeanMitchell Would you ever attend IASP?

[Chris Kennedy @c3K]

@DrSeanMitchell Very cool. What conferences might you present this work at?

[Chris Kennedy @c3K]

@DrSeanMitchell Awesome, maybe we can collaborate on something one of these years. So for suicide risk psychometrics, is that like Obegi 2025 suicidality severity, or more like suicidal ideation assessments? Any Rasch, or no?

[Chris Kennedy @c3K]

@DrSeanMitchell What kind of SRA angles are you interested in out of curiosity?

[Chris Kennedy @c3K]

@RexDouglass That would be awesome

[Chris Kennedy @c3K]

@RexDouglass Would you ever write a perspective piece on this? Or create such a preprint system?

[Chris Kennedy @c3K]

@ben_fifield @ryancbriggs @DrEmmaZang @AdamBerinsky How is it more stringent than an R package version?