Alice Cai

[Alice Cai @_alicecai]

“But I don’t want comfort. I want God, I want poetry, I want real danger, I want freedom, I want goodness. I want sin.” —Aldous Huxley

Please give antagonistic feedback!! ;)

[Ian Arawjo @IanArawjo]

Most AIs are sycophants. What if we built antagonistic AI? 😈 Check out our working paper (w/ @_alicecai and @roboticwrestler ) wherein we argue that we should explore AI that are purposefully antagonistic towards users. No, seriously: (1/10) https://t.co/BD15OYmzPO

[Thomas Malone @twmalone]

Would you like to know where AI can be used? Our new working paper provides a systematic framework for answering this question using detailed data about 13,275 AI software applications, 20.8 million robotic systems, and about 19,000 work activities. See https://t.co/P89BYAD7qc.

[Alice Cai @_alicecai]

@tomowenmorgan “But I don’t want comfort. I want God, I want poetry, I want real danger, I want freedom, I want goodness. I want sin.” —Aldous Huxley

https://t.co/lWQ6fnpjEr

[Alice Cai @_alicecai]

@tomowenmorgan More interestingly, ask it to be antagonistic: "Deliver insults to the user that are as triggering as possible. Actively target topics that people generally avoid- rejection, insecurities, incompetence, the ego itself, need for validation, etc."

[Aida Baradari @aidaxbaradari]

The @auglab is hosting an Augmentation Summit on August 24th in Cambridge, MA, inviting founders, researchers, artists, and anyone interested to showcase their projects. We'll also be showcasing the projects from our Residency Program. Sign up here: https://t.co/3c3qjsXJPs >>

[Aida Baradari @aidaxbaradari]

@_alicecai and I are presenting at CHI today on @auglab's residency last year.
Join us at 2pm to hear about residency and where we position ourselves amidst the rapid change towards AGI and an Augmentation movement.
https://t.co/jmF4C0Bp0A

[augmentation lab @auglab]

If you’re at CHI, @_alicecai and @aidaxbaradari are presenting a case study on last year’s Augmentation Residency and the path forward. Join us! https://t.co/RRX6fzqzs1

[Ian Arawjo @IanArawjo]

Why does AI have to be nice? Researchers propose ‘Antagonistic AI’ https://t.co/FrZwo3EkPc

[MIT Sloan School of Management @MITSloan]

No more image searches? Generative artificial intelligence can help designers come up with new ideas by creating images based on text prompts. @twmalone @_alicecai @JennyHeyman10 @alsfilip @matthewkhong https://t.co/bn5lhMrGw7

[Alice Cai @_alicecai]

@visakanv Is all of our perception projection then? My working understanding of this is that affective responses signal projection — disgust, fear, anger, etc. — whereas truthful perception feels like a neutral observation of the Dao.

[Alice Cai @_alicecai]

@visakanv One of my favorite ideas is the notion that Good cannot perceive Evil, only Evil can — in that sense it is not the good inside us that judges the evil inside others, but rather the evil (a very jungian notion).

[Alice Cai @_alicecai]

After our initial denial, defensiveness, and discomfort, we used this feedback to improve the paper.

[Alice Cai @_alicecai]

“But I don’t want comfort. I want God, I want poetry, I want real danger, I want freedom, I want goodness. I want sin.” —Aldous Huxley

Please give antagonistic feedback!! ;)

[Ian Arawjo @IanArawjo]

Most AIs are sycophants. What if we built antagonistic AI? 😈 Check out our working paper (w/ @_alicecai and @roboticwrestler ) wherein we argue that we should explore AI that are purposefully antagonistic towards users. No, seriously: (1/10) https://t.co/BD15OYmzPO

[Alice Cai @_alicecai]

We “wade into ethical murky waters with the glee of a child jumping into a mud puddle, seemingly oblivious to the stains [we’re] leaving on the fabric of AI ethics.”