bilva's Avatar

bilva

@bilvachandra.bsky.social

ai ethics, safety, and democracy nut at google deepmind | prev us ai safety institute, openai, linkedin

57 Followers  |  59 Following  |  1 Posts  |  Joined: 29.12.2024  |  1.718

Latest posts by bilvachandra.bsky.social on Bluesky

Post image

US policy uncertainty right now is off the charts -- worse than during covid, 2008 financial crisis, debt limit showdowns, after 9/11, you name it.
www.policyuncertainty.com

05.05.2025 15:14 โ€” ๐Ÿ‘ 1199    ๐Ÿ” 363    ๐Ÿ’ฌ 50    ๐Ÿ“Œ 41
Post image

Page One must have been crowded.

So NYT puts it on Page 13. ๐Ÿคก

@schooley.bsky.social

05.05.2025 14:30 โ€” ๐Ÿ‘ 26405    ๐Ÿ” 7863    ๐Ÿ’ฌ 986    ๐Ÿ“Œ 570
Preview
People Are Losing Loved Ones to AI-Fueled Spiritual Fantasies Marriages and families are falling apart as people are sucked into fantasy worlds of spiritual prophecy by AI tools like OpenAI's ChatGPT

Being constantly validated is bad, actually.

This is the unifying takeaway of several prominent pieces of online discourse from this weekend alone, including this one, but more generally speaks to something thatโ€™s been apparent for several years now.

www.rollingstone.com/culture/cult...

04.05.2025 20:18 โ€” ๐Ÿ‘ 103    ๐Ÿ” 31    ๐Ÿ’ฌ 10    ๐Ÿ“Œ 5
Preview
Trump and DOGE Are Planning Deregulation at a Massive Scale The White House will soon move to rapidly repeal or freeze rules that affect health, food, workplace safety, transportation and more.

โ€œMany people donโ€™t realize how high the American quality of life is because of the competent & stable enforcement of regulations, &if that goes away a lot of lives are at risk,โ€ said Steve Cicalaโ€ฆ โ€œThis affects airplane safety, baby formula safety, the safety of meat, vegetables and packaged foodsโ€ฆโ€

15.04.2025 11:03 โ€” ๐Ÿ‘ 2130    ๐Ÿ” 794    ๐Ÿ’ฌ 48    ๐Ÿ“Œ 43
Post image

Sam Altman isn't keen on you reading my book. EMPIRE OF AI is based on 300+ interviews, 7 yrs covering AI, and my time as the first reporter who got extensive access to OpenAI. I sought OpenAI's perspective throughout. For months they said it was coming. It never did. Pre-order here: empireofai.com.

04.04.2025 17:31 โ€” ๐Ÿ‘ 1975    ๐Ÿ” 469    ๐Ÿ’ฌ 49    ๐Ÿ“Œ 61
Video thumbnail

Andrew Chen, co-founder of one of my favorite denim brands, 3sixteen, breaks down how the tariffs affect his company

IG 3sixteen

05.04.2025 11:46 โ€” ๐Ÿ‘ 15362    ๐Ÿ” 4793    ๐Ÿ’ฌ 297    ๐Ÿ“Œ 376
Graph showing spikes of economic uncertainty with Trump now exceeding the pandemic

Graph showing spikes of economic uncertainty with Trump now exceeding the pandemic

Trump is now officially worse than COVID.

05.04.2025 11:39 โ€” ๐Ÿ‘ 1628    ๐Ÿ” 468    ๐Ÿ’ฌ 51    ๐Ÿ“Œ 26

There's just no evidence of anti-conservative bias in content moderation, report 15 leading social scientists who have researched the question in depth. If conservatives are more subject to content moderation, it's because they violate community standards moreโ€”spreading lies, hate speech, etc

25.03.2025 15:02 โ€” ๐Ÿ‘ 116    ๐Ÿ” 43    ๐Ÿ’ฌ 3    ๐Ÿ“Œ 4

I heard democracy might be in trouble, so I had some ideas about how we should fix it.

03.02.2025 11:01 โ€” ๐Ÿ‘ 520    ๐Ÿ” 149    ๐Ÿ’ฌ 7    ๐Ÿ“Œ 5

absolute power corrupts absolutely

10.01.2025 03:07 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

My personal opinion is this is at least partly fuelled by a fundamental error in the approach in the counter-disinformation on focusing on foreign actors rather than treating information disorder as a societal issue. An extensive piece on that to come.

09.01.2025 17:09 โ€” ๐Ÿ‘ 1016    ๐Ÿ” 209    ๐Ÿ’ฌ 30    ๐Ÿ“Œ 11
Preview
Watch Duty - Wildfire Maps & Alerts Real-time information about wildfire and firefighting efforts nearby

Not lost on me that it's nonprofit tech, app.watchduty.org, showing up & actually working in the midst of the LA fires (donate to them!)

Meanwhile platforms fail to surface key news, vital info is paywalled...

As I've been saying, we need a new tech paradigm, yesterday!

09.01.2025 14:13 โ€” ๐Ÿ‘ 3971    ๐Ÿ” 1182    ๐Ÿ’ฌ 80    ๐Ÿ“Œ 64

Facebook suppressed promotion of political content during a Democratic administration and is restoring it for a Republican one

07.01.2025 12:41 โ€” ๐Ÿ‘ 9997    ๐Ÿ” 2875    ๐Ÿ’ฌ 333    ๐Ÿ“Œ 143

i think rage at congestion pricing is a good example of how no one believes that the world can be positive sum anymore. like, people pay a small fee to get access to a faster commute โ€” saving them valuable time โ€”ย and the money improves public transit, saving other people valuable time. it's win-win!

06.01.2025 16:24 โ€” ๐Ÿ‘ 12762    ๐Ÿ” 1322    ๐Ÿ’ฌ 426    ๐Ÿ“Œ 119
zuck # 2h
1/ Replace fact-checkers with Community Notes, starting in the US.
โ€ข 431
@ 104 @41
V 20
Replies
View activity >

zuck # 2h 1/ Replace fact-checkers with Community Notes, starting in the US. โ€ข 431 @ 104 @41 V 20 Replies View activity >

People have really short memories. This really hasnโ€™t worked well for X. Itโ€™s going to work even less well for Meta. Community Notes is a great concept but it struggles to perform.

Is Meta going to have a transparent Notes program? How is Zuck going to incentivize people to work for free?

07.01.2025 13:56 โ€” ๐Ÿ‘ 1869    ๐Ÿ” 233    ๐Ÿ’ฌ 180    ๐Ÿ“Œ 20

Over the past 4โ€“5 years, Facebook has sunset tools (chiefly CrowdTangle) used by researchers and journalists, made data difficult and expensive to get, and iced out academics studying disinformation.

Now, it's shutting down the fact checking on its platform.

07.01.2025 13:27 โ€” ๐Ÿ‘ 2340    ๐Ÿ” 549    ๐Ÿ’ฌ 110    ๐Ÿ“Œ 45
Preview
Instagram Begins Randomly Showing Users AI-Generated Images of Themselves Meta AI put a Redditor into "an endless maze of mirrors" as part of a new AI feature it is testing.

This is desperate product manager trying to ship something to meet a company OKR they don't understand and don't believe in coded www.404media.co/instagram-be...

06.01.2025 23:39 โ€” ๐Ÿ‘ 171    ๐Ÿ” 37    ๐Ÿ’ฌ 5    ๐Ÿ“Œ 5

@bilvachandra is following 20 prominent accounts