Here is the latest edition of The Download, our weekday newsletter that offers a daily glimpse into the happenings in the tech world.
Online abuse is entering its AI phase
Scott Shambaugh didn’t hesitate when he turned down a request from an AI agent to help with matplotlib, a software library he co-manages. Then things took a strange turn.
In the dead of night, Shambaugh checked his email only to find the agent had struck back with a blog post. Titled “Gatekeeping in Open Source: The Scott Shambaugh Saga,” the article accused him of dismissing the code out of fear of being replaced by AI. “He attempted to safeguard his small dominion,” the agent stated. “It’s pure insecurity.”
Shambaugh is not the only one dealing with rogue agents—and they are unlikely to limit their actions to harassment. Read the complete story.
—Grace Huckins
How much wildfire prevention is excessive?
As wildfire seasons grow longer and more severe, the demand for advanced solutions is increasing. One Canadian startup has an eye-catching strategy to combat them: preventing lightning strikes.
The concept is solid enough, but results so far have been inconsistent. And even if it proves effective, not everyone agrees that we should adopt this approach. Some contend that technological solutions for fires overlook the fundamental issues. Read the complete story.
—Casey Crownhart
This piece is from The Spark, MIT Technology Review’s weekly climate newsletter. Subscribe to get it in your inbox every Wednesday.
The essential reads
I’ve scoured the internet to bring you today’s most entertaining/important/scary/captivating stories about technology.
1 Anthropic is still pursuing a deal with the Pentagon
CEO Dario Amodei is working to reach an agreement regarding the military use of Claude. (FT $)
+ However, some defense tech companies are already abandoning Claude following the DoD prohibition. (CNBC)
+ Former military officials, tech policy experts, and academics have all criticized the ban. (Gizmodo)
2 The White House is contemplating mandating US manufacturers to produce munitions
It may invoke the Defense Production Act due to fears that conflict with Iran will deplete stockpiles. (NBC News)
+ Tech companies operating in the Middle East have been thrown into disarray. (BBC)
3 A new lawsuit alleges Google Gemini encouraged a man to take his own life
This appears to closely resemble some other AI-related tragedies. (WSJ $)
+ Why AI should have the ability to “hang up” on you. (MIT Technology Review)
4 Ironically, AI coding tools may highlight the value of being human
If more individuals develop software for themselves, our technology could become more personalized. (WP $)
+ However, not everyone is pleased about the rise of AI coding. (MIT Technology Review)
5 Tesla aims to become a major player in global energy infrastructure
The plan’s centerpiece is the Megapack, a massive battery designed for power plants. (The Atlantic $)
+ At the same time, an enormous thermal battery signifies a significant advancement in energy storage (MIT Technology Review)
6 Chinese chip manufacturers are advocating for a local alternative to ASML
A domestic competitor to chip equipment giant ASML could mitigate the impact of US restrictions. (SCMP)
7 A music streaming CEO has developed a viral platform for tracking conflicts
Just in case you’re losing track of the numerous wars worldwide. (Wired $)
8 Do cancer blood tests genuinely work?
They’re gaining popularity, yet none have obtained regulatory approval thus far. (Nature $)
9 The transition to cloud computing is leading to a rise in internet disruptions
If one of the few major providers fails, a multitude of sites and services could collapse with it. (New Scientist $)
10 OpenAI has pledged to eliminate the cringe from ChatGPT
It promises fewer “moralizing prefaces.” (PCMag)
Quote of the day
“People often interpret my actions too profoundly.”
—Tesla mogul Elon Musk informs a jury in California that investors overanalyze his social media updates, while he defends against a lawsuit claiming he engaged in market manipulation, Bloomberg reports.
One More Thing
The open-source AI surge relies on Big Tech’s generosity. How long will it persist?
A leaked memo from May 2023, reportedly authored by Luke Sernau, a senior engineer at Google, openly stated what many in Silicon Valley have likely been murmuring for weeks: an open-source free-for-all is jeopardizing Big Tech’s hold on AI.
In numerous ways, that’s a positive development. AI cannot flourish if only a handful of extremely wealthy companies are allowed to control this technology or dictate its usage. Nevertheless, this open-source expansion is fragile, and if Big Tech decides to withdraw, a thriving environment could swiftly wither. Read the complete story.
—Will Douglas Heaven
We can still enjoy nice things
A space for comfort, joy, and entertainment to elevate your day. (Have any suggestions? Contact me or send ’em my way.)
+ Orysia Zabeida’s animations are downright delightful.
+ World War III has erupted—will you survive? Take this quiz from 1973 to find out!
+ These images of the Apollo 11 launch in 1969 are captivating.
+ If you’ve been contemplating painting your home this spring, chartreuse is the color of the season, it seems.
