Shortly after its inauguration, the Trump administration has made no secret that it isn't especially interested in funding research. Before January's end, major science agencies had instituted pauses on research funding, and grant funding has not been restored to previous levels since. Many individual grants have been targeted on ideological grounds, and agencies like the National Science Foundation are expected to see significant cuts. Since then, individual universities have been targeted, starting with an ongoing fight with Columbia University over $400 million in research funding.

This week, however, it appears that the targeting of university research has entered overdrive, with multiple announcements of funding freezes targeting several universities. Should these last for any considerable amount of time, they will likely cripple research at the targeted universities.

On Wednesday, Science learned that the National Institutes of Health has frozen all of its research funding to Columbia, despite the university agreeing to steps previously demanded by the administration and the resignation of its acting president. In 2024, Columbia had received nearly $700 million in grants from the NIH, with the money largely going to the university's prestigious medical and public health schools.

Read full article

Comments

© Bruce Yuanyue Bi

In the last week, the US Space Force awarded SpaceX a $5.9 billion deal to make Elon Musk's space company the Pentagon's leading launch provider, and then it assigned the vast majority of this year's most lucrative launch contracts to SpaceX.

On top of these actions, the Space Force reassigned the launch of a GPS navigation satellite from United Launch Alliance's long-delayed Vulcan rocket to fly on SpaceX's Falcon 9. ULA, a joint venture between Boeing and Lockheed Martin, is SpaceX's chief US rival in the market for military satellite launches.

Given the close relationship between Musk and President Donald Trump, it's not out of bounds to ask why SpaceX is racking up so many wins. Some plans floated by the Trump administration involving SpaceX in recent months have raised concerns over conflicts of interest.

Read full article

Comments

© Photo by Mandel Ngan/AFP

Health experts took to one of the country's leading medical journals to pen searing rebukes of Robert F. Kennedy Jr.'s first weeks as the country's top health official—and they called upon their colleagues to rise up to fight the misinformation and distrust they allege Kennedy, a long-time anti-vaccine advocate, is fomenting.

From gutting federal health agencies and knee-capping critical local public health programs, to delaying a significant vaccine advisory meeting, hiring a discredited anti-vaccine advocate to conduct a vaccine study, ousting the country's top vaccine regulator, and undermining the response to the mushrooming measles outbreak in Texas that stands to threaten the country's measles elimination status—the researchers had no shortage of complaints.

In one article, pediatric infectious disease expert Kathryn Edwards of Vanderbilt University recounted the timeline of the measles outbreak, noting the missteps, missed opportunities, and controversial comments Kennedy made along the way. The rundown included his trivialization of the outbreak, failure to strongly advocate for vaccination, promotion of unproven treatments, like cod liver oil, and delayed responses from the Centers for Disease Control and Prevention, which Kennedy controls.

Read full article

Comments

© Getty | Jim Watson

Google made waves with the release of Gemini 2.5 last month, rocketing to the top of the AI leaderboard after previously struggling to keep up with the likes of OpenAI. That first experimental model was just the beginning. Google is deploying its improved AI in more places across its ecosystem, from the developer-centric Vertex AI to the consumer Gemini app.

Gemini models have been dropping so quickly, it can be hard to grasp Google's intended lineup. Things are becoming clearer now that the company is beginning to move its products to the new branch. At the Google Cloud Next conference, it has announced initial availability of Gemini 2.5 Flash. This model is based on the same code as Gemini 2.5 Pro, but it's faster and cheaper to run.

You won't see Gemini 2.5 Flash in the Gemini app just yet—it's starting out in the Vertex AI development platform. The experimental wide release of Pro helped Google gather data and see how people interacted with the new model, and that has helped inform the development of 2.5 Flash.

Read full article

Comments

© Thomas Fuller/SOPA Images/LightRocket via Getty Images

WASHINGTON, DC—Over the course of a nearly three-hour committee hearing Wednesday, the nominee to lead NASA for the Trump administration faced difficult questions from US senators who sought commitments to specific projects.

However, maneuvering like a pilot with more than 7,000 hours in jets and ex-military aircraft, entrepreneur and private astronaut Jared Isaacman dodged most of their questions and would not be pinned down. His basic message to members of the Senate Committee on Commerce, Science, and Transportation was that NASA is an exceptional agency that does the impossible, but that it also faces some challenges. NASA, he said, receives an “extraordinary” budget, and he vowed to put taxpayer dollars to efficient use in exploring the universe and retaining the nation’s lead on geopolitical competitors in space.

“I have lived the American dream, and I owe this nation a great debt,” said Isaacman, who founded his first business at 16 in his parents' basement and would go on to found an online payments company, Shift4, that would make him a billionaire. Isaacman is also an avid pilot who self-funded and led two private missions to orbit on Crew Dragon. Leading NASA would be “the privilege of a lifetime,” he said.

Read full article

Comments

© NASA/Bill Ingalls

An anti-deepfake bill is on the verge of becoming US law despite concerns from civil liberties groups that it could be used by President Trump and others to censor speech that has nothing to do with the intent of the bill.

The bill is called the Tools to Address Known Exploitation by Immobilizing Technological Deepfakes On Websites and Networks Act, or Take It Down Act. The Senate version co-sponsored by Ted Cruz (R-Texas) and Amy Klobuchar (D-Minn.) was approved in the Senate by unanimous consent in February and is nearing passage in the House. The House Committee on Energy and Commerce approved the bill in a 49-1 vote yesterday, sending it to the House floor.

The bill pertains to "nonconsensual intimate visual depictions," including both authentic photos shared without consent and forgeries produced by artificial intelligence or other technological means. Publishing intimate images of adults without consent could be punished by a fine and up to two years of prison. Publishing intimate images of minors under 18 could be punished with a fine or up to three years in prison.

Read full article

Comments

© Getty Images

On Wednesday, Donald Trump, once again, took to Truth Social to abruptly shift US trade policy, announcing a 90-day pause "substantially" lowering reciprocal tariffs against all countries except China to 10 percent.

Because China retaliated—raising tariffs on US imports to 84 percent on Wednesday—Trump increased tariffs on China imports to 125 percent "effective immediately." That likely will not be received well by China, which advised the Trump administration to cancel all China tariffs Wednesday, NPR reported.

"The US's practice of escalating tariffs on China is a mistake on top of a mistake," the Chinese finance ministry said, calling for Trump to "properly resolve differences with China through equal dialogue on the basis of mutual respect."

Read full article

Comments

© Aurich Lawson | Getty Images

Spammers used OpenAI to generate messages that were unique to each recipient, allowing them to bypass spam-detection filters and blast unwanted messages to more than 80,000 websites in four months, researchers said Wednesday.

The finding, documented in a post published by security firm SentinelOne’s SentinelLabs, underscores the double-edged sword wielded by large language models. The same thing that makes them useful for benign tasks—the breadth of data available to them and their ability to use it to generate content at scale—can often be used in malicious activities just as easily. OpenAI revoked the spammers’ account after receiving SentinelLabs’ disclosure, but the four months the activity went unnoticed shows how enforcement is often reactive rather than proactive.

“You are a helpful assistant”

The spam blast is the work of AkiraBot—a framework that automates the sending of messages in large quantities to promote shady search optimization services to small- and medium-size websites. AkiraBot used python-based scripts to rotate the domain names advertised in the messages. It also used OpenAI’s chat API tied to the model gpt-4o-mini to generate unique messages customized to each site it spammed, a technique that likely helped it bypass filters that look for and block identical content sent to large numbers of sites. The messages are delivered through contact forms and live chat widgets embedded into the targeted websites.

Read full article

Comments

© Getty Images | Iurii Motov

On Wednesday, Anthropic introduced a new $100- to $200-per-month subscription tier called Claude Max that offers expanded usage limits for its Claude AI assistant. The new plan arrives after many existing Claude subscribers complained of hitting rate limits frequently.

"The top request from our most active users has been expanded Claude access," wrote Anthropic in a news release. A brief stroll through user feedback on Reddit seems to confirm that sentiment, showing that many Claude users have been unhappy with Anthropic's usage limits over the past year—even on the Claude Pro plan, which costs $20 a month.

One of the downsides of a relatively large context window with Claude (the amount of text it can process at once) has been that long conversations or inclusions of many reference documents (such as code files) fill up usage limits quickly. That's because each time the user adds to the conversation, the entire text of the conversation (including any attached documents) is fed back into the AI model again and re-evaluated. But on the other hand, a large context window allows Claude to process more complex projects within each session.

Read full article

Comments

© Anthropic

Some elements of Microsoft's Copilot assistant in Windows 11 have felt like a solution in search of a problem—and it hasn't helped that Microsoft has frequently changed Copilot's capabilities, turning it from a native Windows app into a web app and back again.

But I find myself intrigued by a new addition to Copilot Vision that Microsoft began rolling out this week to testers in its Windows Insider program. Copilot Vision launched late last year as a feature that could look at pages in the Microsoft Edge browser and answer questions based on those pages' contents. The new Vision update extends that capability to any app window, allowing you to ask Copilot not just about the contents of a document but also about the user interface of the app itself.

Microsoft's Copilot Vision update can see the contents of any app window you share with it. Credit: Microsoft

Provided the app works as intended—not a given for any software, but especially for AI features—Copilot Vision could replace "frantic Googling" as a way to learn how to use a new app or how to do something new or obscure in complex PC apps like Word, Excel, or Photoshop. I recently switched from Photoshop to Affinity Photo, for example, and I'm still finding myself tripped up by small differences in workflows and UI between the two apps. Copilot Vision could, in theory, ease that sort of transition.

Read full article

Comments

© Microsoft

A rogue cyborg security (SEC) unit gains autonomy and must learn to interact with humans while hiding its new capability in the trailer for Murderbot, Apple TV+'s new 10-episode sci-fi comedic thriller starring Alexander Skarsgård. It's based on the bestselling book series The Murderbot Diaries by Martha Wells. And judging from the trailer, Murderbot looks like it will strike just the right balance between humor and action.

There are seven books in Wells' series thus far. All are narrated by Murderbot, who is technically owned by a megacorporation but manages to hack and override its governor module. Rather than rising up and killing its former masters, Murderbot just goes about performing its security work, relieving the boredom by watching a lot of entertainment media.

In the first book, All Systems Red, Murderbot saves a scientific expedition on an alien planet when they are attacked by a giant alien creature. During the ensuing investigation, the cyborg uncovers a plot against the expedition, as well as a second team, by yet another team intent on killing their rivals for some reason. Murderbot joins the humans in foiling those murderous plans but escapes onto a cargo ship at the end rather than give up its hard-earned autonomy.

Read full article

Comments

© Apple TV+

Last week, Nintendo made the unprecedented move of delaying US Switch 2 preorders to "assess" the impact of Donald Trump's massive tariffs on the countries where the console is produced. Before most of those tariffs were recently delayed for 90 days, many were wondering if the company may be mulling a last-minute increase in the Switch 2's $450 asking price to account for those import taxes.

While industry analysts think that kind of immediate price increase is unlikely, they warn that Trump's tariffs could have longer-term impacts on Switch 2 pricing and supplies in the US for years to come.

Already baked in

DFC Intelligence CEO David Cole, for instance, said in a recent analyst note that the company is currently modeling "a 20 percent price increase over the next two years" across all video game hardware thanks to "broader macroeconomic challenges." In the case of the Switch 2, though, Cole clarified that "we believe much of the 20 percent increase was already baked into the $450 price," which Nintendo is "not likely" to raise at this point.

Read full article

Comments

© Nintendo

Later today, a former Facebook employee, Sarah Wynn-Williams, will testify to Congress that Meta executives "repeatedly" sought to "undermine US national security and betray American values" in "secret" efforts to "win favor with Beijing and build an $18 billion dollar business in China."

In her prepared remarks, which will be delivered at a Senate subcommittee on crime and counterterrorism hearing this afternoon, Wynn-Williams accused Meta of working "hand in glove" with the Chinese Communist Party (CCP). That partnership allegedly included efforts to "construct and test custom-built censorship tools that silenced and censored their critics" as well as provide the CCP with "access to Meta user data—including that of Americans."

Wynn-Williams worked as Facebook's Director of Global Public Policy from 2011 to 2017. She left at the height of the Cambridge Analytica scandal, just before Mark Zuckerberg got grilled by Congress over misinformation and election interference on its platform.

Read full article

Comments

© Bloomberg / Contributor | Bloomberg

Even the tiniest of living things are capable of some amazing forms of locomotion, and some come with highly sophisticated sensor suites and manage to source their energy from the environment. Attempts to approach this sort of flexibility with robotics have taken two forms. One involves making tiny robots modeled on animal behavior. The other involves converting a living creature into a robot. So far, either approach has involved giving up a lot. You're either only implementing a few of life's features in the robot or shutting off most of life's features when taking over an insect.

But a team of researchers at Harvard has recognized that there are some behaviors that are so instinctual that it's possible to induce animals to act as if they were robotic. Or mostly robotic, at least—the fruit flies the researchers used would occasionally go their own way, despite strong inducements to stay with the program.

Smell the light

The first bit of behavior involved Drosophila's response to moving visual stimuli. If placed in an area where the fly would see a visual pattern that rotates from left to right, the fly will turn to the right in an attempt to keep the pattern stable. This allowed a projector system to "steer" the flies as they walked across an enclosure (despite their names, fruit flies tend to spend a lot of their time walking). By rotating the pattern back and forth, the researchers could steer the flies between two locations in the enclosure with about 94 percent accuracy.

Read full article

Comments

© arlindo71

Google has unveiled a new AI processor, the seventh generation of its custom TPU architecture. The chip, known as Ironwood, was reportedly designed for the emerging needs of Google's most powerful Gemini models, like simulated reasoning, which Google prefers to call "thinking." The company claims this chip represents a major shift that will unlock more powerful agentic AI capabilities. Google calls this the "age of inference."

Whenever Google talks about the capabilities of a new Gemini version, it notes that the model's capabilities are tied not only to the code but to Google's infrastructure. Its custom AI hardware is a key element of accelerating inference and expanding context windows. With Ironwood, Google says it has its most scalable and powerful TPU yet, which will allow AI to act on behalf of a user to proactively gather data and generate outputs. This is what Google means when it talks about agentic AI.

Ironwood delivers higher throughput compared to previous Google Tensor Processing Units (TPUs), and Google really plans to pack these chips in. Ironwood is designed to operate in clusters of up to 9,216 liquid-cooled chips, which will communicate directly with each other through a newly enhanced Inter-Chip Interconnect (ICI).

Read full article

Comments

© Google

A rare spot of good news today: For the second year in a row, US roads got a little safer. The National Highway Traffic Safety Administration published its early estimate of road deaths in 2024; 39,345 people lost their lives, which is a 3.8 percent decrease from the 40,901 deaths that occurred on US roads in 2023.

The problem started with the pandemic; although road traffic dried up, the death rate leapt by 20 percent.

There's no single cause, and studies have identified multiple contributing factors: empty roads designed to practically encourage speeding, little to no enforcement of traffic laws by the police, a general sense of fatalism in the face of public health restrictions that few Americans had ever contemplated in recent times, and car companies making big trucks and SUVs with high hoods, which are much more deadly to pedestrians and other vulnerable road users in a crash.

Read full article

Comments

© Getty Images

As President Donald Trump signed a slew of executive orders Tuesday aimed at keeping coal power alive in the United States, he repeatedly blamed his predecessor, Democrats, and environmental regulations for the industry’s dramatic contraction over the past two decades.

But across the country, state and local officials and electric grid operators have been confronting a factor in coal’s demise that is not easily addressed with the stroke of a pen: its cost.

For example, Maryland’s only remaining coal generating station, Talen Energy’s 1.3-gigawatt Brandon Shores plant, will be staying open beyond its previously planned June 1 shutdown, under a deal that regional grid operator PJM brokered earlier this year with the company, state officials, and the Sierra Club.

Read full article

Comments

© US Global Change

Are you a millennial parent who has made cycling your entire personality but have found it socially unacceptable to abandon your family for six hours on a Saturday? Or are you a bike-curious urban dweller who hasn’t owned a bicycle since middle school? Do you stare at the gridlock on your commute, longing for a bike-based alternative, but curse the errands you need to run on the way home?

I have a solution for you: invest in a cargo bike.

Cargo bikes aren't for everyone, but they're great if you enjoy biking and occasionally need to haul more than a bag or basket can carry (including kids and pets). In this guide, we'll give you some parameters for your search—and provide some good talking points to get a spouse on board.

Read full article

Comments

© Aurich Lawson | John Timmer

Last week, Nintendo made waves across the game industry by announcing that Mario Kart World would sell for a suggested price of $80 in the US. That nominal price represents a new high-water mark both for Nintendo and for the game industry at large, which has generally reserved prices above $70 for fancy, trinket-laden collectors' editions or Digital Deluxe Editions that include all variety of downloadable bonuses.

When you adjust historical game prices for inflation, though, you find that asking $80 for a baseline game in 2025 is broadly in line with the prices big games were commanding 10 to 15 years ago. And given the faster-than-normal inflation rates of the last five years, even the $70 nominal game prices that set a new standard in 2020 don't have the same purchasing oomph they once did.

The data

To judge Mario Kart World's $80 price against historical trends, we first needed to figure out how much games cost in the past. To do that, we built off of our similar 2020 analysis, which relied on scanned catalogs and retail advertising fliers for real examples of nominal console game pricing going back to the Atari era. For more recent years, we relied more on press reports and archived digital storefronts to show what prices new games were actually selling for at the time.

Read full article

Comments

© Nintendo

A leak of 190,000 chat messages traded among members of the Black Basta ransomware group shows that it’s a highly structured and mostly efficient organization staffed by personnel with expertise in various specialties, including exploit development, infrastructure optimization, social engineering, and more.

The trove of records was first posted to file-sharing site MEGA. The messages, which were sent from September 2023 to September 2024, were later posted to Telegram in February 2025. ExploitWhispers, the online persona who took credit for the leak, also provided commentary and context for understanding the communications. The identity of the person or persons behind ExploitWhispers remains unknown. Last month’s leak coincided with the unexplained outage of the Black Basta site on the dark web, which has remained down ever since.

“We need to exploit as soon as possible”

Researchers from security firm Trustwave’s SpiderLabs pored through the messages, which were written in Russian, and published a brief blog summary and a more detailed review of the messages on Tuesday.

Read full article

Comments

© Getty Images