Read

Read

895 bookmarks
Custom sorting
Displaced Los Angeles-area residents face spiking rents as authorities warn of price gouging
Displaced Los Angeles-area residents face spiking rents as authorities warn of price gouging
The ongoing wildfires have set off a stampede to find housing, leading some homeowners and property managers to raise rents higher than the 10% allowed during emergencies.
In some cases, prospective short-term renters in Los Angeles County are finding themselves on waiting lists, in bidding wars or being asked to provide a year’s worth of rent up front.
·nbcnews.com·
Displaced Los Angeles-area residents face spiking rents as authorities warn of price gouging
DeepSeek-R1’s bold bet on reinforcement learning: How it outpaced OpenAI at 3% of the cost
DeepSeek-R1’s bold bet on reinforcement learning: How it outpaced OpenAI at 3% of the cost
DeepSeek-R1’s Monday release has sent shockwaves through the AI community, disrupting assumptions about what’s required to achieve cutting-edge AI performance. This story focuses on exactly how DeepSeek managed this feat, and what it means for the vast number of users of AI models. For enterprises developing AI-driven solutions, DeepSeek’s breakthrough challenges assumptions of OpenAI's dominance -- and offers a blueprint for cost-efficient innovation.
·venturebeat.com·
DeepSeek-R1’s bold bet on reinforcement learning: How it outpaced OpenAI at 3% of the cost
DeepSeek FAQ
DeepSeek FAQ
I don't buy that any of this is a path to AGI but it's certainly disruptive.
OpenAI’s gambit for control — enforced by the U.S. government — has utterly failed.
·stratechery.com·
DeepSeek FAQ
I knew one day I’d have to watch powerful men burn the world down – I just didn’t expect them to be such losers | Rebecca Shaw
I knew one day I’d have to watch powerful men burn the world down – I just didn’t expect them to be such losers | Rebecca Shaw
I don't appreciate the last line conflating these guys with nerds! But I guess that's another thing they've taken.
I knew that one day we might have to watch as capitalism and greed and bigotry led to a world where powerful men, deserving or not, would burn it all down. What I didn’t expect, and don’t think I could have foreseen, is how incredibly cringe it would all be.
while you may be able to buy power, it’s impossible to buy a good personality
This combination of evil and embarrassment is a unique horror, one that science fiction has failed to prepare us for.
It’s time for us to start getting revenge on the nerds.
·theguardian.com·
I knew one day I’d have to watch powerful men burn the world down – I just didn’t expect them to be such losers | Rebecca Shaw
The Slop Society
The Slop Society
I barely use the bad platforms but his final points about everything getting worse because the ad revenue gets prioritized for growth
It's a comfortable lie to say that Meta has "suddenly" done something, because it gives the media (and society at large) air cover for ignoring a gaping wound in the internet.
It's almost ridiculous to call us "users" of Facebook at this point. We are the used, the punished, the terrorized, the tricked, the scammed, and the abused, constantly forced to navigate through layers of abstraction between the thing we are allegedly using and the features we'd like to use.
so many products on the App Store have expensive and exploitative microtransactions, because Apple makes 30% off of all App Store revenue, even if the app sucks.
Trump is perhaps the most transactional human being ever to grace the earth.
I fear that the subtle little problems you see every day will both multiply and expand, and that the core services we use will break down, because I believe the most powerful in big tech never really gave a shit and no longer believe they have to pretend otherwise.
·wheresyoured.at·
The Slop Society
Microsoft AI Red Team says security work will never be done
Microsoft AI Red Team says security work will never be done
Of course not, you have untrusted inputs and analyzing them practically requires another model that also makes mistakes.
Due to fundamental limitations of language models, one must assume that if an LLM is supplied with untrusted input, it will produce arbitrary output. When that input includes private information, one must also assume that the model will output private information.
·theregister.com·
Microsoft AI Red Team says security work will never be done