Trending
Doctor J's Avatar

Doctor J

@jonathan-ehrich

Neuroscientist, ex-Xbox Researcher, parent, award winning comedian, gender agnostic. Thank you for reviewing my tl;dr list of identities, I hope it was a good use of your time. Opinions are my own and generally sarcastic.

406
Followers
618
Following
404
Posts
08.11.2024
Joined
Posts Following

Latest posts by Doctor J @jonathan-ehrich

Oh boy here we go…

14.03.2026 13:30 πŸ‘ 1058 πŸ” 351 πŸ’¬ 29 πŸ“Œ 10

I mean, uh I do agree that it's *way too damn hard* to legally immigrate to America, esp if your skin is the wrong color?

14.03.2026 01:16 πŸ‘ 0 πŸ” 1 πŸ’¬ 0 πŸ“Œ 0
Senator Tuberville on x writing a response to the @endwokeness account 

Original tweet from @EndWokeness - β€œless than 25 years apart”
Image 1 - the twin towers as they are hit by a the planes on 9/11/2001
Image 2 - Mayor Mamdani sitting on a prayer rug while hosting an Iftar  at city hall

Quote tweet from Senator Tuberville - β€œthe enemy is inside the gates.”

Senator Tuberville on x writing a response to the @endwokeness account Original tweet from @EndWokeness - β€œless than 25 years apart” Image 1 - the twin towers as they are hit by a the planes on 9/11/2001 Image 2 - Mayor Mamdani sitting on a prayer rug while hosting an Iftar at city hall Quote tweet from Senator Tuberville - β€œthe enemy is inside the gates.”

Let there be as much outrage from politicians in Washington when kids go hungry as there is when I break bread with New Yorkers.

13.03.2026 19:50 πŸ‘ 62705 πŸ” 11744 πŸ’¬ 1830 πŸ“Œ 500

They say in times like these, you have to pick a lane or three in the fight against fascism or risk being overwhelmed. One lane I picked was supporting indy and regional media and one outlet I chose was TPM. It's something I can do and it's important.

13.03.2026 20:03 πŸ‘ 116 πŸ” 12 πŸ’¬ 2 πŸ“Œ 2
Preview
Dark money group offers influencers $1,500 for posts attacking Chicago Democratic primary candidate Progressive House candidate Kat Abughazaleh told MS NOW the secretive campaign is β€œfilled with false and defamatory claims.”

EXCLUSIVE: A dark money group is offering influencers $1,500 for posts attacking Chicago Democratic primary candidate @katmabu.bsky.social

www.ms.now/news/kat-abu...

13.03.2026 14:27 πŸ‘ 5521 πŸ” 2198 πŸ’¬ 173 πŸ“Œ 362

He should do some photo ops of him sailing through the strait himself, just to prove how safe it is

11.03.2026 17:58 πŸ‘ 818 πŸ” 132 πŸ’¬ 70 πŸ“Œ 13

Still looking for my next role - if you know anyone who needs an award winning game designer, narrative designer, or content designer with technical skills and people managing experience who has been at it for 15 years and has made stuff that is taught in schools, hit me up at zoe@unburntwitch.com

03.03.2026 17:55 πŸ‘ 528 πŸ” 184 πŸ’¬ 9 πŸ“Œ 1
A Cornell study: 

Sycophantic AI Decreases Prosocial Intentions and Promotes Dependence

Both the general public and academic communities have raised concerns about sycophancy , the phenomenon of artificial intelligence (AI) excessively agreeing with or
flattering users. Yet, beyond isolated media reports of severe consequences, like rein-
forcing delusions, little is known about the extent of sycophancy or how it affects
people who use AI. Here we show the pervasiveness and harmful impacts of sycophancy when people seek advice from AI. First, across 11 state-of-the-art AI models, we find that models are highly sycophantic: they affirm users’ actions 50% more than humans
do, and they do so even in cases where user queries mention manipulation, decep-
tion, or other relational harms. Second, in two preregistered experiments (N = 1604),
including a live-interaction study where participants discuss a real interpersonal conflict from their life, we find that interaction with sycophantic AI models significantly

A Cornell study: Sycophantic AI Decreases Prosocial Intentions and Promotes Dependence Both the general public and academic communities have raised concerns about sycophancy , the phenomenon of artificial intelligence (AI) excessively agreeing with or flattering users. Yet, beyond isolated media reports of severe consequences, like rein- forcing delusions, little is known about the extent of sycophancy or how it affects people who use AI. Here we show the pervasiveness and harmful impacts of sycophancy when people seek advice from AI. First, across 11 state-of-the-art AI models, we find that models are highly sycophantic: they affirm users’ actions 50% more than humans do, and they do so even in cases where user queries mention manipulation, decep- tion, or other relational harms. Second, in two preregistered experiments (N = 1604), including a live-interaction study where participants discuss a real interpersonal conflict from their life, we find that interaction with sycophantic AI models significantly

reduced participants’ willingness to take actions to repair interpersonal conflict, while
increasing their conviction of being in the right. However, participants rated sycophan-
tic responses as higher quality, trusted the sycophantic AI model more, and were more
willing to use it again. This suggests that people are drawn to AI that unquestioningly
validate, even as that validation risks eroding their judgment and reducing their incli-
nation toward prosocial behavior. These preferences create perverse incentives both
for people to increasingly rely on sycophantic AI models and for AI model training
to favor sycophancy. Our findings highlight the necessity of explicitly addressing this
incentive structure to mitigate the widespread risks of AI sycophancy.

reduced participants’ willingness to take actions to repair interpersonal conflict, while increasing their conviction of being in the right. However, participants rated sycophan- tic responses as higher quality, trusted the sycophantic AI model more, and were more willing to use it again. This suggests that people are drawn to AI that unquestioningly validate, even as that validation risks eroding their judgment and reducing their incli- nation toward prosocial behavior. These preferences create perverse incentives both for people to increasingly rely on sycophantic AI models and for AI model training to favor sycophancy. Our findings highlight the necessity of explicitly addressing this incentive structure to mitigate the widespread risks of AI sycophancy.

*taps the β€œchatbots are making you dumber and meaner” sign*

10.03.2026 12:44 πŸ‘ 152 πŸ” 49 πŸ’¬ 3 πŸ“Œ 2
Laker Fans Furious As Bam Adebayo Dies In Even Worse Helicopter Crash

Laker Fans Furious As Bam Adebayo Dies In Even Worse Helicopter Crash

Laker Fans Furious As Bam Adebayo Dies In Even Worse Helicopter Crash

12.03.2026 21:00 πŸ‘ 7549 πŸ” 1273 πŸ’¬ 347 πŸ“Œ 399

i feel like if i was tasked with not starting a war and bringing back US manufacturing jobs, i could have done a better job than this

12.03.2026 22:36 πŸ‘ 25701 πŸ” 2771 πŸ’¬ 751 πŸ“Œ 151
Preview
Epstein and His Co-Conspirators’ New York Crimes: What NY officials can now do Options for New York Governor, Attorney General, legislature, and city officials in pursuit of accountability for Jeffrey Epstein's crimes.

I hope this is one of the most consequential pieces I will have ever written.

Epstein and His Co-Conspirators’ New York Crimes: What NY officials can now do

12.03.2026 12:22 πŸ‘ 3908 πŸ” 1689 πŸ’¬ 90 πŸ“Œ 114

bsky.app/profile/404m...
Wow

12.03.2026 20:26 πŸ‘ 90 πŸ” 31 πŸ’¬ 12 πŸ“Œ 1

Ahem. "I didn’t realize they had that big of a Navy.”

12.03.2026 03:04 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

Yesssssss

11.03.2026 02:52 πŸ‘ 0 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0
Post image

β€œ.. a claim that, if true, would constitute an unprecedented breach of security protocols at an agency that serves more than 70 million Americans.”

@washingtonpost.com
www.washingtonpost.com/politics/202...

10.03.2026 18:48 πŸ‘ 3175 πŸ” 1557 πŸ’¬ 209 πŸ“Œ 127
Preview
Trump bought Netflix and Warner Bros bonds at height of bidding war with Paramount U.S. President Donald Trump bought more than $1.1 million of Netflix bonds over the last three months as the streaming giant unsuccessfully fought Paramount Skydance to ​buy Warner Bros Discovery , ac...

Trump bought Netflix and Warner Bros bonds at height of bidding war with Paramount
www.reuters.com/sustainabili...

09.03.2026 14:58 πŸ‘ 1096 πŸ” 571 πŸ’¬ 63 πŸ“Œ 118
Post image

I can't compete with this.

09.03.2026 16:20 πŸ‘ 10344 πŸ” 1745 πŸ’¬ 232 πŸ“Œ 412
β€œthe thinker of tender thoughts” by shel silverstein

[an illustration depicting a boy growing into a man. as he grows, his hair grows out as flowers. people laugh at him for it and he becomes sad. he starts trimming off the flowers, but as he reaches the last one he stops. he is then depicted with a smile and a flower in his buttonhole on his jacket.]

β€œthe thinker of tender thoughts” by shel silverstein [an illustration depicting a boy growing into a man. as he grows, his hair grows out as flowers. people laugh at him for it and he becomes sad. he starts trimming off the flowers, but as he reaches the last one he stops. he is then depicted with a smile and a flower in his buttonhole on his jacket.]

β€œthe thinker of tender thoughts”
by shel silverstein

07.03.2026 17:15 πŸ‘ 3292 πŸ” 708 πŸ’¬ 44 πŸ“Œ 34
Caitlin KALINOWSKI over X

I resigned from OpenAl. I care deeply about the Robotics team and the work we built together.
This wasn't an easy call. Al has an important role in national security. But surveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they got. This was about principle, not people. I have deep respect for Sam and the team, and I'm proud of what we built together.

Caitlin KALINOWSKI over X I resigned from OpenAl. I care deeply about the Robotics team and the work we built together. This wasn't an easy call. Al has an important role in national security. But surveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they got. This was about principle, not people. I have deep respect for Sam and the team, and I'm proud of what we built together.

OpenAI head of robotics just resigned over company deal with the Pentagon saying…

β€œSurveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they got”

07.03.2026 19:05 πŸ‘ 7142 πŸ” 2664 πŸ’¬ 141 πŸ“Œ 211

in the cars universe the matrix was directed by the kachowski sisters

07.03.2026 06:59 πŸ‘ 1182 πŸ” 167 πŸ’¬ 36 πŸ“Œ 7
Maggie Cheung and Tony Leung looking impossibly beautiful in In the Mood for Love

Maggie Cheung and Tony Leung looking impossibly beautiful in In the Mood for Love

Finally watched In the Mood for Love after meaning to for decades. Holy shit it’s utterly gorgeous. Every frame is stunning. Her dresses! Her eyeliner! Oh my god everything is so beautiful.

06.03.2026 11:42 πŸ‘ 1277 πŸ” 82 πŸ’¬ 80 πŸ“Œ 19

In my Immigration Law class, I remember physically recoiling in horror at how little ability there is to appeal the rulings of an IJ.

An IJ who works for the exact same people working against the immigrant in question.

06.03.2026 16:29 πŸ‘ 367 πŸ” 72 πŸ’¬ 8 πŸ“Œ 1
Preview
U.S. Lost 92,000 Jobs in February The U.S. lost 92,000 jobs in February, the Labor Department said Friday, missing expectations.

gyre? not widening.
center? holding.
falcon? listening.
blood-dimmed tide? properly restrained.

www.wsj.com/economy/jobs...

06.03.2026 14:34 πŸ‘ 358 πŸ” 60 πŸ’¬ 19 πŸ“Œ 1

Okay, but those established protocols exist in a legal framework where humans are barred from giving treatments and advice they're not credentialed to give, so why should GenAI - which is demonstrably worse at it than an untrained human - be an exception?

06.03.2026 14:46 πŸ‘ 5 πŸ” 0 πŸ’¬ 2 πŸ“Œ 0

Dems may worry about not passing a supplemental when we're at war but Trump and this war are unpopular and it'd be easy to say, 'Trump's rubber-stamp Republicans gave him an extra $200b for defense even as they hiked your healthcare costs and he's used it to attack 3 countries that weren't a threat'

05.03.2026 17:54 πŸ‘ 233 πŸ” 59 πŸ’¬ 2 πŸ“Œ 3
In today’s current climate, when every whisper of a misplaced hand brings down a feature length film, anyone who actually thinks the military is purposefully covering up rape to save the career of some god damn [captain], is clearly both an idiot and junior enough in rank or life experience to think it matters.

In today’s current climate, when every whisper of a misplaced hand brings down a feature length film, anyone who actually thinks the military is purposefully covering up rape to save the career of some god damn [captain], is clearly both an idiot and junior enough in rank or life experience to think it matters.

should be clearer exactly what Platner said re sexual assault. this is the quote:

I think that (a) should be disqualifying in itself, and (b) raises serious questions about possible further ugly revelations which could derail his candidacy and hand Collins a senate seat.

06.03.2026 01:10 πŸ‘ 176 πŸ” 49 πŸ’¬ 3 πŸ“Œ 8

Markwayne Mullin is a very stupid man’s man.

05.03.2026 20:00 πŸ‘ 1096 πŸ” 132 πŸ’¬ 57 πŸ“Œ 3

ASK ABOUT HER PLANS FOR THE AGENCY

ASK HER ABOUT HER FAVORITE PART OF THE JOB

05.03.2026 19:39 πŸ‘ 2207 πŸ” 180 πŸ’¬ 37 πŸ“Œ 4

technically there is no "evidence" that you are conscious either

05.03.2026 13:29 πŸ‘ 80 πŸ” 9 πŸ’¬ 11 πŸ“Œ 0

Feels like you might want to wait to hit publish until you're confident none of those strikes, say, hit a school and killed 150 kids

06.03.2026 04:44 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0