Security teams log 54% of successful attacks and alert on just 14%. The rest move through your environment unseen.
The Picus whitepaper shows how breach and attack simulation tests your SIEM and EDR rules so threats stop slipping by detection.
Researchers identified what they believe is the first documented case of a ransomware operation, JadePuffer, conducted entirely by a large language model (LLM) agent.
According to cloud security company Sysdig, JadePuffer used an autonomous AI agent for reconnaissance on the target, to steal credentials, move laterally, establish persistence, escalate privileges, and to encrypt data.
The researchers say that the AI agent adapted to failures during the intrusion, much like a human operator would handle obstacles.
“The operation also adapted in real time, retrying failed steps within refined parameters. In one sequence, it went from a failed login to a working fix in 31 seconds,” Sysdig says.
JadePuffer gained initial access to the target by exploiting CVE-2025-3248, an unauthenticated remote code execution vulnerability in Langflow, a popular open-source framework used for building LLM apps.
The vendor fixed the flaw on April 1, 2025, and in early May of the same year, CISA tagged it as exploited in attacks targeting internet-exposed endpoints, usually deployed with minimal hardening but containing cloud credentials and API keys.
After obtaining code execution through CVE-2025-3248, the AI agent dumped Langflow’s PostgreSQL database, collected host information, searched for environment variables and sensitive files, retrieved credentials, and enumerated a MinIO object store.
Sysdig highlights the adaptive approach to MinIO enumeration, where if one API request returned XML instead of JSON, the next payload adjusted its parsing logic accordingly.
JadePuffer also established persistence on the Langflow host by installing a cron job on the server, which was configured to beacon to the attacker’s infrastructure every 30 minutes.
From the Langflow instance, the attacker pivoted to a production MySQL server running Alibaba Nacos (Naming and Configuration Service), using root credentials whose origin Sysdig couldn’t determine.
Nacos was targeted with multiple payloads, including one exploiting CVE-2021-29441, an authentication bypass vulnerability that creates rogue administrator accounts.
The agent probed for container escape methods and deployed the ransomware payload. According to the researchers, JadePuffer encrypted 1,342 Nacos service configuration items before deleting the originals.
“The captured payloads show the agent encrypting all 1,342 Nacos service configuration items using MySQL’s AES_ENCRYPT(), dropping the original config_info and history tables, and creating an extortion table (README_RANSOM) containing the demand, a Bitcoin payment address, and a Proton Mail contact,” describes Sysdig.

The ransom note claims that the data was encrypted using the AES-256 algorithm, although the researchers believe this to be an overstatement, and that the use of the weaker AES-128-ECB is more likely.
Sysdig mentions that the encryption key is randomly generated but not stored or transmitted to the attacker.
The Bitcoin address listed in the ransom note is an example address widely used in public documentation, possibly the result of the LLM reproducing it from the training data.
Other signs that AI was controlling the attack include detailed natural-language comments in the generated code describing operational reasoning and rapid attack iteration that considers the specific errors encountered, rather than being simple retries.

Sysdig concludes that the case of JadePuffer demonstrates that the age of “agentic threat actors” (ATAs) has arrived, lowering the skill required for conducting damaging cyberattacks.
At the same time, given how AI agents operate today, LLM-generated payloads create new detection opportunities for security solutions.
Security teams log 54% of successful attacks and alert on just 14%. The rest move through your environment unseen.
The Picus whitepaper shows how breach and attack simulation tests your SIEM and EDR rules so threats stop slipping by detection.
Amid the dozens of brands available at Harbor Freight, a few stand out. Icon is one of the most notable brands in the discount retailer’s current range, with its high quality, mechanic-oriented series of tools having been available since 2018. In recent years, the Icon brand’s lineup has expanded significantly, and several high profile products have gone viral across social media.
You might have seen reviewers testing out the capabilities of the brand’s much-hyped magnetic mat, or perhaps you’ve been drawn in by comparison tests that evaluate the price and performance of Icon’s ratchets against Snap-On. Either way, Icon has certainly made waves with the online community of Harbor Freight enthusiasts.
However, despite that online attention, some of the brand’s latest and greatest products are not actually available to purchase on Harbor Freight’s website. At the time of writing, these five products all remain in-store exclusives, although some look more likely than others to eventually be made available for online purchase.
Browse through the range of electrical pliers on Harbor Freight’s website and you’ll find plenty of products that can be added straight to your cart, but the Icon 6 inch flush cut pliers are not among them. At the time of writing, these $34.99 pliers are instead only available in-store. Like all the Icon products here, it’s still worth checking Harbor Freight’s website to see stock levels at your nearest retail location before you go, since availability may be limited in certain locations.
The pliers feature a ¾-inch jaw capacity, and can therefore handle a variety of different sizes of wires, zip ties, or anything else you’ll need to cut through. Their grips are made from a nonslip material, while the blades are made from heat-treated steel. Icon covers all of its hand tools with a limited lifetime warranty, which covers any defects in either the material or the construction of the tool.
Icon makes no secret of the fact that it benchmarks its products against leading tool truck brands, and in some cases, it also takes heavy inspiration from their product names to boot. The Icon T10 diagnostic scanner is the Harbor Freight brand’s version of the Snap-On Triton-D10, but as you might expect, it’s far cheaper. Icon sells its scanner for $1,699.99, while Snap-On’s equivalent offering carries an MSRP of $6,550.
Buyers of the Icon scanner also receive a free year of Icon’s proprietary TrueFix diagnostics software, although after that, a subscription is required. The scanner comes with various accessories, including a USB borescope inspection camera, a wireless 12V battery tester, and OBD-II and DOIP cables. According to the brand, the scanner’s 6,300mAh battery should provide up to 8 hours of runtime before it’ll need to be put back into its docking station. A one year warranty is included as standard.
Although a significant number of Icon’s G2 ratchets are available to purchase online, some of its latest launches remain in-store exclusives for now. At the time of writing, the brand’s ½-inch drive, 12 inch standard ratchet with Comfort Grip is a brand new addition to the lineup, and so it’s not yet available to purchase online.
Harbor Freight first introduced the Icon G2 ratchets in 2025, and there are a few key differences between its latest line and the older G1 line. G2 ratchets are designed with sealed heads and feature gears made from nickel-chromium-molybdenum alloy, making them more durable than before. Icon’s selection of different sizes and types of ratchet has continually expanded since the launch of the G2 line, and the latest drop in June 2026 adds a further 15 different variants to the range.
One thing that hasn’t changed is their competitive pricing, with Icon’s G2 ratchets being much cheaper than their Snap-On rivals. The ½-inch drive, 12 inch standard ratchet with the comfort grip handle retails for $54.99. As a bonus, all of the G2 ratchet line is covered by a lifetime warranty.
With its prominent “Icon” logo embroidered into the cushion, the Icon heavy duty mechanics roller seat looks more premium than similar seats from fellow in-house Harbor Freight brand Pittsburgh. It’ll hold more weight too, with Pittsburgh’s roller seat able to accommodate a maximum of 250 lbs while Icon’s seat has a capacity of 350 lbs.
Then, there are the smaller details: The Icon’s integrated parts tray features dividers to keep small items and essential tools within easy reach, while the Pittsburgh’s tray is just one flat surface. Add in the additional tool drawer that slots neatly under the seat cushion, and it’s easy to see why the Icon seat commands a higher price than the Pittsburgh.
Whether that price is worth it is another question, and will depend on how often you plan on using the seat. The Icon seat retails for $74.99, which is over double the price of its budget rival. However, it’s still far cheaper than a similar seat from a tool truck brand like Snap-On. The Icon seat also remains an in-store exclusive for now, while the Pittsburgh seat is available to purchase straight from the retailer’s website.
Another Icon tool that’s designed to rival Snap-On is the ⅜-inch drive, 5 to 100 ft-lb digital torque wrench. It features an LCD screen and an indicator light to let users know at a glance when the desired torque has been applied. It also offers the option to add up to nine preset torque settings into memory.
In an independent test by YouTuber Project Farm, the torque wrench performed impressively well compared to its tool truck competitor. Although it wasn’t quite as accurate as the Snap-on tool, the difference was relatively small given the significant price difference between the two. When Project Farm tested the tool in 2025, it retailed for $379, but as of June 2026, Harbor Freight has dropped its price to $359.99.
According to its maker, the torque wrench should be accurate between +/- 2% clockwise and +/- 3% counterclockwise. The tool only needs to swing 5° to start tightening, and its flex-head adjusts 15° for optimal versatility. Readouts are available in ft-lb, in-lb, Nm, Kgcm, and dNm. It’s another Icon tool that promises professional-grade accuracy without the tool truck price tag, but it’s only available in Harbor Freight stores, and not online for now.
Looking for the most recent Connections answers? Click here for today’s Connections hints, as well as our daily answers and hints for The New York Times Mini Crossword, Wordle, Connections: Sports Edition and Strands puzzles.
Today’s NYT Connections puzzle contains a fun Looney Tunes-related category, which I thought was hilarious. Read on for clues and today’s Connections answers.
The Times has a Connections Bot, like the one for Wordle. Go there after you play to receive a numeric score and to have the program analyze your answers. Players who are registered with the Times Games section can now nerd out by following their progress, including the number of puzzles completed, win rate, number of times they nabbed a perfect score and their win streak.
Read more: Hints, Tips and Strategies to Help You Win at NYT Connections Every Time
Here are four hints for the groupings in today’s Connections puzzle, ranked from the easiest yellow group to the tough (and sometimes bizarre) purple group.
Yellow group hint: Oh my gosh.
Green group hint: Classic kid projects.
Blue group hint: Road Runner: Beep-beep!
Purple group hint: Swipe right.
Yellow group: Stunning news.
Green group: Science Fair model subjects.
Blue group: ACME products used by Wile E. Coyote.
Purple group: Starting with dating apps.
Read more: Wordle Cheat Sheet: Here Are the Most Popular Letters Used in English Words
The completed NYT Connections puzzle for July 6, 2026.
The theme is stunning news. The four answers are bombshell, revelation, shocker and thunderbolt.
The theme is science fair model subjects. The four answers are atom, DNA, solar system and volcano.
The theme is ACME products used by Wile E. Coyote. The four answers are earthquake pills, iron bird seed, rocket skates and TNT.
The theme is starting with dating apps. The four answers are bumblebee, grind rail, matcha and tinderbox.
While the device has a questionable future, and Elon Musk has denied the report, SpaceX is said to be taking on Apple by shifting into AI hardware, reportedly showing investors a prototype before the company’s IPO.

SpaceX’s artificial intelligence arm, xAI, has been working on a considerably more grounded product it wants to sell to consumers, as it competes against Apple Intelligence and other AI platforms. One that doesn’t involve being blasted off the planet.
According to the Wall Street Journal, Elon Musk’s rocket company has worked on a prototype for an AI device for some time. It was shown off to investors and other stakeholders before the company’s IPO.
Continue Reading on AppleInsider | Discuss on our Forums

JBL packed a color touchscreen into the charging case of the Live Beam 3 earbuds, priced at $99.95 (was $149.95). That addition changes how people interact with their audio gear during a normal day. Someone heading to work or hitting the gym can open the case and swipe through options right there. Volume goes up or down with a touch. Playback pauses or skips without pulling out a phone. ANC modes switch from full noise blocking to letting in some surroundings. Even EQ presets become accessible on the go.
A tap wakes up the display, which shows the battery levels for both the buds and the casing at a glance. Users can pick a favorite photo as the background, and it rotates appropriately when the lid opens, allowing guests to see it right side up. A flashlight option brightens the screen for quick light in a pinch. Shortcuts remain adjustable in the companion app, so only the most often used tools appear. This setup addresses a major complaint about many wireless earphones. Touch controls on the buds handle basic commands but frequently need sacrifices. Adjusting one thing results in losing simple access to another. The case screen eliminates this limitation and keeps everything in one easy location that goes with the buds anyway.
The sound quality backs up the convenience, as these earbuds provide a dynamic presentation with plenty of detail and strong bass that sounds engaging across a variety of music styles. Listeners that prefer a little more vitality in their songs will find it enjoyable. Advanced codec compatibility enables compatible phones and players to send higher-quality audio when available. Those who wish to fine-tune their EQ can use an app.
The battery life is outstanding, with a single charge lasting nine to ten hours with noise canceling turned on. If you turn off ANC, it will last about twelve hours. The enclosure increases the total playback time to nearly two full days of use. When time is of the essence, a quick ten-minute charge via USB-C adds an extra four hours.
Silicone tips provide fit by forming a seal that is beneficial to both sound quality and noise reduction. The buds’ IP55 rating ensures that sweat from exercises or unexpected rain will not be an issue. Many individuals find the stem style comfortable for lengthy usage, however results vary depending on ear shape, as with most in-ear designs. Active noise canceling works nicely in this price range. It successfully minimizes traffic noise, train noise, and workplace chatter, making it suitable for everyday commutes or concentrated work. Adaptive modifications are made automatically based on the surroundings and fit. Calls come through clearly thanks to multiple beamforming microphones that eliminate wind and background interference. Multipoint Bluetooth allows the buds to connect to two devices at once, making it easy to swap between a phone and a laptop or tablet.
It’s not often we mention the Kardashian-Jenner clan here at Trusted Reviews, but Kylie Jenner’s surprise collaboration with Meta is all I’ve been thinking about.
In case you missed it, the youngest Jenner recently unveiled her own pair of Meta smart glasses. Coined Starfire, the oval-shaped specs are not only framed as a trendy choice but they’re fitted with Meta’s controversial features, including Meta AI and, most notably, the built-in 12MP camera.
Kylie’s collaboration with Meta is surprising and disappointing for so many reasons. Firstly, in a viral interview back in May, she recalled how scary and invasive growing up with paparazzi essentially stalking her for photos was.
Like her eldest sister, she’s known for keeping certain parts of her life private. For example, she hid her first pregnancy entirely from the media, and then later was reluctant to share photos of her second child online. This is completely understandable, as everyone has a right to privacy and absolutely shouldn’t feel any need to share images online.
With the above in mind, why on earth is Kylie therefore promoting smart glasses that have the power to take privacy away from pretty much anyone who has the bad luck of walking past a desperate aspiring content-creator-slash-creep?
Meta glasses have a terrible reputation for being a complete privacy nightmare, especially when it comes to women and girls’ safety. Back in May, the BBC reported that a woman was going about her day when a man approached her, without a camera or phone in hand. Instead, he was wearing smart glasses, and she had “no idea she was being filmed”.
The video was then posted online and viewed thousands of times, with the woman only finding out when a friend sent it to her.
While we don’t know the exact brand of smart glasses the man was wearing, Meta glasses all have a light that comes on when you’re filming, which technically should show people that they’re being filmed. However, and I’ve seen this for myself, that light literally couldn’t be any smaller. I would totally understand if someone passed off the light as a simple reflection or maybe even just a large scratch on the glasses.


That’s not the only worrying story. As uncovered by Wired last month, Meta has recently embedded face-recognition technology into the Meta AI app. While it’s not currently accessible by users, it will identify people captured by the glasses’ camera and alert the wearer when it recognises someone.
This has, unsurprisingly, caused concern. Experts who spoke to The Independent earlier this year feared this technology could pose a “direct and serious risk” to domestic abuse survivors as it could enable their abusers to locate and track (in other words, stalk) them, without them even knowing.
Plus, the fact that anyone who walks past a Meta glasses wearer’s image will be “cropped, indexed, and saved to a folder marked ‘pending’” is incredibly unnerving. What if your neighbour or the fellow commuter who always gets the same train as you is wearing Metas? Will your image be consistently stored for them to see? Will Meta actually note that you’re a frequent passer-by and attempt to identify you?
I’ve had hands-on experience with both the Meta Ray-Bans 2 and the Oakley Meta Vanguard too. The latter I somewhat understand the purpose of more, as they’re used as sports glasses and enable you to capture your surroundings, get real-time stats and more without needing to reach for your phone. The Ray-Bans 2 and similar glasses, on the other hand, are a different story.


Yes, Meta glasses allow you to do interesting things like translate live, but so do AirPods and many of the best Android phones. And yes, the glasses also give you real-time answers with Meta AI without you needing to reach for your phone, but is anything really that urgent?
I admit, I just can’t get on board with smart glasses, and maybe it’s because I’m not the target audience. But once you factor in the high price, the limited style options and, most notably, the serious privacy concerns, the cons surely vastly outweigh the pros.
However, Kylie Jenner’s influence is undeniable, and Meta clearly knows this as she’s one of the most followed users on Instagram. Her collaboration with Meta is not only hypocritical from someone who publicly states how much she favours privacy, but it will undoubtedly attract a new demographic of younger users who grow to think it’s simply fine to film people without their knowledge.
Obviously (and very unfortunately) it’s not as easy to say “let’s just ban smart glasses”, but there undoubtedly needs to be more regulation of filming and sharing content online.
Looking for a different day?
A new Quordle puzzle appears at midnight each day for your time zone – which means that some people are always playing ‘today’s game’ while others are playing ‘yesterday’s’. If you’re looking for Sunday’s puzzle instead then click here: Quordle hints and answers for Sunday, July 5 (game #1623).
Quordle was one of the original Wordle alternatives and is still going strong now more than 1,500 games later. It offers a genuine challenge, though, so read on if you need some Quordle hints today — or scroll down further for the answers.
Enjoy playing word games? You can also check out my NYT Connections today and NYT Strands today pages for hints and answers for those puzzles, while Marc’s Wordle today column covers the original viral word game.
SPOILER WARNING: Information about Quordle today is below, so don’t read on if you don’t want to know the answers.
• The number of different vowels in Quordle today is 4*.
* Note that by vowel we mean the five standard vowels (A, E, I, O, U), not Y (which is sometimes counted as a vowel too).
• The number of Quordle answers containing a repeated letter today is 0.
• No. None of Q, Z, X or J appear among today’s Quordle answers.
• The number of today’s Quordle answers starting with the same letter is 2.
If you just want to know the answers at this stage, simply scroll down. If you’re not ready yet then here’s one more clue to make things a lot easier:
• T
• S
• P
• S
Right, the answers are below, so DO NOT SCROLL ANY FURTHER IF YOU DON’T WANT TO SEE THEM.
The answers to today’s Quordle, game #1624, are…
This game worked out very well for me, although that was largely due to a quartet of pretty straightforward words.
TRAWL was my only moment of doubt but felt right as I slowly tapped the letters out.
The answers to today’s Quordle Daily Sequence, game #1624, are…
In the mid-noughties, when music by the Killers and Franz Ferdinand blared out of every pub and nightclub I passed, I spent my days and nights struggling through a Ph.D. in applied mathematics. My research focused on simulating how special light waves interact in liquid crystals and using simple equations to approximate and understand those interactions. When I look back at my thesis now, liquid crystal technology is old hat, and I imagine my work could be completed with AI assistance in a matter of days—maybe hours.
But the same cannot be said for the work of the pure mathematics Ph.D. students with whom I shared a cramped office at the University of Edinburgh. At the time, I felt sorry for these colleagues, who day after day sat at their desks, seemingly tearing their hair out and making no progress. (Though I was struggling too, I was at least always making some headway.) When we finished and went our separate ways, some hadn’t even published a paper.
Now, in hindsight, I finally understand why they toiled for years on abstract mathematical problems that only a handful of people in the world care about. It wasn’t arrogance, as I thought at the time; they weren’t trying to prove their superior intelligence by being the first to solve a seemingly intractable mathematical problem. It wasn’t even a form of masochism (which was my second guess)—penance for some imagined inadequacy. I realized they derived joy, satisfaction, and meaning from the long journey toward understanding.
“Sometimes, understanding just strikes you as being very beautiful. Sometimes it’s a feeling of accomplishment, like completing a marathon,” muses Carnegie Mellon University mathematician Jeremy Avigad. “But it’s not quite either of those: It’s just a wonderful feeling when you’ve been thinking long and hard about something complex, difficult, and then—all of a sudden—it just comes together.”
This feeling has driven mathematicians throughout history. Likewise, the way mathematicians pursue that feeling has changed little over the centuries. They notice or imagine links, patterns, or properties in numbers, shapes, or logical structures. From this, they write conjectures—unproven statements of their speculation. They or other mathematicians then use logical reasoning and the tools of mathematics in often creative ways to prove or disprove those conjectures. Finally, yet other mathematicians verify (or challenge) the proofs.
Invariably, this process requires a whole heap of thinking time. “I went to a pure maths camp with classes where we would sit with hard maths problems for half an hour and no one would say anything—everyone was just thinking,” says Krystal Maughan, a mathematician and computer scientist about to get her Ph.D. at the University of Vermont. “But then we would work together and kind of tease out the problem.”
This is the age-old joy of math in action. But today’s AI systems are starting to make inroads into bypassing this slow, deliberative process. Taking this trend to its logical conclusion, what happens if AI makes the mathematician’s struggle completely unnecessary? Might AI even sideline humanity completely?
For decades, computation has accelerated mathematical progress. This began 50 years ago, when mathematicians used a computer to prove the four-color theorem, which asks whether any map can be colored using no more than four colors, with no adjacent regions sharing the same color. The answer is yes, and the computer proved it, controversially, by checking 1,936 cases in a way no human could realistically verify.
Yet throughout this computational era, even in proofs relying on massive computational resources, the role of the human mathematician has remained central. Humans propose conjectures, guided by intuition. They devise strategies to prove them, guided by creativity and experience. And humans verify whether those proofs are correct.
Now AI is challenging the status quo. In just a few years, large language models (LLMs) have evolved from “stochastic parrots,” capable of little more than regurgitating basic mathematics scraped from the internet, into advanced mathematical reasoning machines.
Last summer, systems from Google DeepMind and OpenAI reached a level equivalent to the world’s most mathematically gifted high school students, achieving gold-medal status at the International Mathematical Olympiad. In this annual competition, contestants must solve six notoriously difficult problems from various areas of mathematics.
Earlier this year, Google DeepMind’s experimental AI system Aletheia achieved an even more significant milestone when it autonomously produced publishable Ph.D.-level research results. While the work itself is obscure mathematically—calculating structure constants in arithmetic geometry—the significance lies in the complex reasoning it displayed in tackling an unsolved mathematical problem. And more recently, a new general-purpose AI system from OpenAI disproved an important conjecture in combinatorial geometry. This result would have been worthy of publication in a major mathematics journal if humans had been the authors, and top mathematicians hailed the feat as a milestone for AI in mathematics, demonstrating independent, original, and sophisticated thinking.
Another shift has come from combining LLMs with mathematical tools known as proof assistants, which have been around for more than a decade. These systems—such as Isabelle, Lean, and Rocq—are specialized programming languages that check mathematical proofs step-by-step, verifying their logical correctness. Traditionally, mathematicians have had to translate their theorems and proofs into this machine-readable format by hand, a laborious process known as formalization. Now, LLMs are starting to remove this bottleneck, automating the translation of informal proofs into formal code that proof assistants can verify.
Versions of such systems, sometimes called reasoning agents, are becoming highly sophisticated. In February, for example, the AI company Math, Inc. used its aspirationally named reasoning agent Gauss to formalize a proof that had earned the mathematician Maryna Viazovska, of EPFL, in Switzerland, a Fields Medal in 2022. Gauss first helped human mathematicians complete the formalization of Viazovska’s solution to the 8-dimensional sphere-packing problem in a matter of days, and then autonomously formalized the more complicated 24-dimensional case in just two weeks.
Such achievements suggest that AI is already capable of handling some mathematical tasks long considered uniquely human. As the technology advances, more of the day-to-day work of human mathematicians is likely to become fair game for AI.

Gluekit
Human mathematicians could become “priests to oracles.” —Yang-Hui He, London Institute for Mathematical Sciences
In September 2025, I attended the 12th Heidelberg Laureate Forum—an annual conference that brings hundreds of young mathematicians and computer scientists together with their intellectual idols. AI dominated the conversation and, from the get-go, tension was in the air.
Speakers described a future in which superhuman AI mathematicians transcend human knowledge and capabilities: forming conjectures, searching solution spaces, proving conjectures, and finally verifying the proofs and generalizing the results, all without human involvement. If this future comes to pass, Yang-Hui He of the London Institute for Mathematical Sciences memorably declared, human mathematicians could become “priests to oracles.”
While such startling predictions were being voiced on stage, my gaze was drawn to the audience. Frowning, fidgeting, and exchanging furtive glances—the crowd’s unease was palpable. Trill White, a student at Australia’s Deakin University, later recalled sitting in that hall and thinking: “ ‘That’s devastating. What will people have to contribute to mathematics? Will it become something that no one understands?’ I did get a sense that this is going to change everything.”

Gluekit
“We certainly started realizing AI has the potential to replace us.” —Jessica Randall, Google Developer Groups
Jessica Randall, a South African mathematician for Google Developer Groups, says she sensed a collective existential dread rising among the young mathematicians. “I could feel everyone was worried, because they hadn’t thought that far ahead,” she says. “It was like a big bombshell that hit us, and we certainly started realizing AI has the potential to replace us.”
Some established mathematicians, including He, seem comfortable with AI taking on tasks that are currently the preserve of human mathematicians. That’s because they just want to know the answers to the biggest questions in mathematics—such as the six remaining Millennium Prize Problems—even if AI does it all. “A lot of mathematicians are pragmatic and just want to understand. They would sell their soul for the solution to a problem,” jokes Avigad. “Whatever it takes, right?”
But this “just want to know” camp is by no means the only faction: Most mathematicians do not hope or expect AI to replace them entirely. Instead, two broad alternatives are emerging. The first is a human-centric aspiration that prioritizes human understanding of mathematics and treats AI as a tool, much like a calculator. The second is a collaborative “teamwork makes the dream work” vision, where humans and AI work together to tackle problems neither could solve alone.
Fields Medalist and Princeton mathematician Akshay Venkatesh has been thinking about this topic from the human-centric viewpoint for years. In 2022, he used his Fields Medal Symposium to implore the mathematics community to deeply consider what AI might mean for the practice of mathematics. At the time, the idea that AI could replace mathematicians seemed far-fetched. Now, he says, “we’re reaching the point where, for at least some tasks with abstract mathematical reasoning, computers are becoming competitive with humans.”
For Venkatesh, the question is not just what computers can do, but what mathematics is for. “Sometimes I think when we use numbers, it’s not so much that we are describing phenomena that are intrinsically numerical, but that we can all agree exactly what the numbers mean,” he says. “It’s a way of bringing us to agreement.”

Maia Fraser of the University of Ottawa argues that mathematics is more than finding answers. For her, the struggle to understand a problem is one of the discipline’s greatest rewards.
Markian Lozowchuk
Mathematician and machine learning expert Maia Fraser, of the University of Ottawa, shares this sentiment. She says the joy she derives from mathematics is something distinctly human that integrates the subconscious and conscious mind. She describes starting with an intuitive sense that a certain thing should be true and gradually bringing out something that she can express in a rigorous proof. Communicating and sharing these deep-born thoughts is “a form of collective intelligence that is something beautiful about the human spirit,” she says.
By these arguments, an AI proof of a mathematical conjecture that has stubbornly resisted human efforts would be useful only if comprehensible to humans. “That the statement can be proved by AI is already useful information,” concedes Fraser. “But then it’s still an open problem to come up with an elegant, beautiful human proof.” Even if no such proof exists, she says, searching for it “is still a valuable endeavor.”
A more collaborative approach to AI in mathematics comes from Terence Tao, who first competed in the math Olympiad at the age of 10. In 1986, 1987, and 1988, he won bronze, silver, and gold medals, respectively, making him the youngest winner of each of the three medals in Olympiad history. Now a Fields Medalist and professor at the University of California, Los Angeles, he has earned a reputation as one of the most gifted mathematicians alive.
Unlike some of his peers, Tao is neither dismissive of AI nor fearful. Instead, he sees it as the catalyst for a fundamental shift in the discipline—a transition toward what he calls “big mathematics.” He envisions a future of large-scale, decentralized collaborations between humans and machines, where complex mathematical tasks can be diced and sliced, with humans claiming the creative parts and AI doing the lion’s share of the technical grunt work.
Already, Tao is experimenting with this concept, working on problems alongside scores of online collaborators, some using AI tools. “A hundred years ago, almost every mathematics paper was single author,” he says. “But now I collaborate with people I’ve never met—and maybe in the future, I won’t even know if they are AI or real people.”
The key to Tao’s vision is uniquely mathematical: formalization. When a proof is translated into code and checked step-by-step by proof assistants, it removes any chance of human error or dishonesty. This approach changes how collaboration works, because trust is established through verification rather than reputation or rapport. An idea from an unknown researcher or even an amateur can be taken seriously if it has a formal proof.
“If it wasn’t for this formal verification layer, opening projects up without any safeguards would just be a disaster,” adds Tao. “But in math, we can completely check and verify outputs, and this really filters out a lot of the rubbish.”
From the young researchers at the Heidelberg Laureate Forum to some of the biggest names in the field, mathematicians all seem to agree on one point: AI has the potential to transform their discipline. But there’s far less consensus on what that transformation will mean in practice.
Some worry about the accessibility of AI tools. Traditionally, mathematicians have required little more than intuition, training, and a pen and paper to advance their field. If this slow, deliberative process is no longer valued by society, and particularly by research funders, then mathematics could become an elitist activity, only practiced by select organizations that can afford to work with proprietary AI models.
Another concern is motivation. As AI systems take on more of the work, the incentive to engage deeply with difficult problems may weaken. Princeton’s Venkatesh says that the long human process of formulating and understanding a proof may be hard to justify, not just to funders, but even to mathematicians themselves. “There have been times where I’ve spent years thinking about something, and I’ve slowly struggled to understand it,” he says. “If your computer can do large chunks of that for you, will you have the motivation to spend that time?”
That concern extends to the next generation. If students can use AI to jump straight to answers, they most likely will. But every time they skip the struggle, they miss an opportunity to build the foundations of their own unique intuition. Over time, some worry, the next generation of mathematicians may suffer from a form of intellectual atrophy, unable to think outside the AI box that trained them.
In response to such fears, the mathematics community is taking action. Individuals are writing essays, organizing workshops, and debating in journals, while institutions and community groups are developing guidelines for how AI should be used in research and publication. Indeed, mathematicians are applying the same rigor and curiosity that they use every day to reckon with the challenges of AI. Taken together, these efforts reflect a broad effort to try to retain control over the direction of mathematics in the era of AI.
So, is AI sucking the soul out of math? In one way, it is doing the opposite. It is forcing mathematicians to confront deep questions about what mathematics is, why they have devoted their lives to it, and the purpose math serves in society. At the same time, though, it is reshaping the practice of mathematics in a way that may be difficult to reverse.
“Mathematics makes me a better problem solver at normal problems, because it frames my mind to think in a very logical, rational way,” says Randall, who noted the existential dread at the Heidelberg Forum. “It helps with every aspect of my life.” As AI transforms mathematics, many researchers wonder whether future mathematicians will be able to say the same.
From Your Site Articles
Related Articles Around the Web

Portable computers in the mid-1980s were finally small enough to carry, yet their screens kept pulling users back toward desks and power outlets. The BBC program Micro Live used a January 1986 episode to lay out exactly why that gap existed and what might close it.
The segment began at the Which Computer Show, where two new approaches were presented side by side. Sharp provided a laptop with a backlit liquid crystal display. The extra light made the image easier to read in normal surroundings, but the underlying LCD still confined users to a narrow sweet zone directly in front of the screen. When I moved slightly to the side, the contrast collapsed. Colors and details simply disappeared. Toshiba displayed a plasma panel beside it. A fine grid of wires spanned inside the screen, illuminating a gas in bright spots when voltage crossed the lines. On camera, the image appeared clear and vivid. In actuality, the design consumed significantly more electricity than a battery could provide for an extended period of time, ran hot, and required frequent refreshing. That refresh cycle produced apparent flicker, which many users previously blamed for tired eyes after extended sessions.
Presenters pointed out the common flaws without drama. LCDs remained cool and consumed little electricity while providing poor contrast and narrow viewing angles. Plasma screens provided more brightness and greater angles in some situations, but they required mains electricity and caused the flutter associated with constant image refreshes. Neither provided the clarity that people expected from paper or typical desk monitors. One presentation summarized the aim that everyone kept missing. What portable users actually required, he claimed, was a screen that remained vast in area while being compact overall, cost little to make, emitted no radiation, remained flicker-free, provided strong contrast, traveled smoothly, and consumed nearly no power.

The program then moved to Harlow’s research laboratory. Engineers had created a functioning prototype that approached the refresh problem from a new perspective. Two sheets of glass were only 11 microns apart. Tiny glass threads kept the gap stable. The tight slot was filled with a unique fluid that required only a few drops to cover the entire panel. Each glass sheet had parallel lines of conductive material. Each crossing of those lines produced a controllable point on the image. The fluid behaved differently based on the electrical signal used. A high-frequency pulse aligned the molecules, allowing light to pass right through. Instead, they were scattered by a low-frequency signal, which blocked or redirected light. Once the molecules had settled into any arrangement, they remained there. No further electrical push was required to hold the image. As a result, the prototype could maintain a stable image even after power had been removed from the panel, something ordinary LCD and plasma panels could not.

Demonstrations made the difference clear, since a typical laptop screen went black when the power was turned off and lost detail as the viewer switched position. The new panel kept its image viewable and the contrast consistent across far broader angles. The effect was more like to a printed page than the flickering electrical displays most people had seen on portable devices. Because the design did not include the polarizing filters that are often required by LCD displays, construction remained easier. Fewer layers resulted in less light loss and potentially lower manufacturing costs. Once an image was saved, power consumption remained minimal because nothing needed to cycle to keep it.

Practical hurdles remained, as each pixel required approximately 200 volts to flip states, which was far more than typical logic voltage. A full-size prototype has hundreds of thousands of discrete connections along its edges. Engineers have already begun gluing special driver chips directly to the glass, reducing the number of external cables.
Nearly 1 million people have lost a total of $3.8 billion after buying President Donald Trump’s $TRUMP memecoin, according to cryptocurrency analytics firm Nansen.
The New York Times reports that Nansen’s analysis is based on transactions that are publicly visible on the blockchain, showing that 988,905 accounts had lost money on the memecoin as of the end of June. That represents around two out of three $TRUMP buyers.
On Sunday, $TRUMP was trading at $1.69, down nearly 98% from its high of $75.35.
Trump announced the memecoin three days before his inauguration in 2025. He’d previously co-founded a crypto startup, World Liberty Financial, with his sons. The $WLFI coin has also declined significantly in value.
In a recent financial disclosure, the president revealed that he made $636 million from the $TRUMP memecoin, accounting for nearly half of the $1.4 billion that the president made from the crypto industry last year.
Under the Trump administration, the Securities and Exchange Commission has said it will not regulate memecoins as securities and has dropped a number of lawsuits against crypto companies. A White House spokesperson told the NYT, “President Trump proudly made the United States the crypto capital of the world.”
![]()
According to the post, each processor combines 6 “Coyote Cove” P-cores, 12 “Arctic Wolf” E-cores, and 4 LP-E cores. That mix suggests a design that balances compute throughput with background and low-power tasks, rather than just piling on more performance cores.
Read Entire Article
Source link
Weekend Open Thread: High Hopes
The House | “Reframing the debate from a binary discussion of winners and losers”: Yuan Yang reviews ‘We Are Not Machines’
Strategy authorizes up to $1.25B in Bitcoin sales under new capital plan
How to Build INSANE Live Financial Dashboards With Claude
Anonymous researcher drops 0-day ‘exploitarium’ repo
Australia treasurer says alleged access of prime minister’s bank data ’incredibly concerning’
Taylor Swift and Travis Kelce wedding staffer hilariously struggles to keep her cool while checking in megastars
The AI boom won’t burst all at once. It will pop in ‘rolling bubbles’: Macquarie
Broncos roster: OL Ben Powers (No. 74) entering final year of contract
Airdrop Registration Becomes Key Focus For Remittix As RTX Launch Updates Approach
Presenter Caroline Flack’s brother Paul Flack dies aged 55
Binance stock trading tops $1B in first month after launch
Alibaba-affiliate Ant Group enters the humanoid robot market with 12 deals
New exhibition reflects five decades of movement between island of Ireland and GB
South Africa proposes crypto tax guidance under existing rules
Standard Chartered Secures MiCA License as ESMA Adds 37 New Crypto Firms
Lenovo laptops are now shipping with YMTC SSDs, a sign of Chinese NAND entering the mainstream
Meta Platforms Stock Jumps 7% Today as Bloomberg Reports Company Plans to Enter the Cloud Business
What a 10 Percent Drop Means for Buyers, Sellers and Renters
Binance Re-Enters Philippines As EU MiCA Rules Restrict Access
You must be logged in to post a comment Login