Ashley Capoot / CNBC: Filing: Elon Musk texted Greg Brockman about settling days before trial; after being denied, he said Brockman and Altman “will be the most hated men in America”  —  Two days before Elon Musk's multi-billion-dollar lawsuit against OpenAI was slated to head to trial …

Solving multiplayer games with function approximation The post Playing Connect Four with Deep Q-Learning appeared first on Towards Data Science.

Onshore wind development in the United States is being brought to a standstill.

Computational model lets users tweak parameters to hear effect on the sound in early design process.

The AI market is full of big promises of grand transformation. Health care is a prime target for those promises, beset as it is by financial pressures, labor shortages, and the growing burden of caring for an aging population. AI developers are targeting functions that vary widely, from curing cancer and performing surgery to streamlining…

Discussion | Link

Seth Fiegerman / Bloomberg: Source: OpenAI has raised over $4B at a $10B pre-money valuation for The Deployment Company, a new joint venture to help businesses adopt OpenAI tools  —  OpenAI has raised more than $4 billion for a new joint venture that will focus on helping businesses adopt its artificial intelligence software …

Table of Contents Semantic Caching for LLMs: TTLs, Confidence, and Cache Safety Why Semantic Caching for LLMs Requires Production Hardening Cache TTL in Semantic Caching: Preventing Stale LLM Responses MLOps Project Structure for Semantic Caching with FastAPI and Redis How… The post Semantic Caching for LLMs: TTLs, Confidence, and Cache Safety appeared first on PyImageSearch.

The hacking prowess of Anthropic's Mythos AI has gotten a lot of attention, including from the NSA. Samuel Boivin/NurPhoto via Getty Images The cybersecurity community went on alert when Anthropic announced on April 7, 2026, that its latest and most capable general-purpose large language model, Claude Mythos Preview, had demonstrated remarkable – and unintended – capabilities. The artifical intelligence system was able to find and exploit software vulnerabilities – the most serious type of software bugs – at a rate not seen before. The news ignited concern among the public, world governments and the information technology sector about the capabilities of today’s AI to undermine cybersecurity, with some people framing the model as a global cybersecurity threat. Claiming that it would be too risky to release the model, and that the company had the moral responsibility to disclose these vulnerabilities, Anthropic said it would not immediately offer the model to the public. Instead, it granted exclusive access to tech giants to test the model’s capabilities, a process Anthropic dubbed Project Glasswing. As a cybersecurity researcher, I think Mythos’ capabilities are impressive, but the AI system does not represent a radical departure. Mythos is less a new threat than a mirror reflecting how people behave and how fragile modern systems already are. What Mythos did During a controlled evaluation, engineers with minimal security experience prompted Mythos to scan thousands of software codebases for vulnerabilities. The model showed striking capabilities in conducting multistep, autonomous attacks that take experts weeks or even months to put together. Mythos was not only able to discover 271 vulnerabilities in Mozilla’s Firefox, it also developed exploits to take advantage of 181 of those. Overall, Anthropic’s red team, which takes on the role of an attacker to test defenses, and the United Kingdom’s AI Security Institute reported that Mythos found thousands of zero-day, or previously unreported, vulnerabilities in major operating systems, web browsers and other applications – software flaws that have not yet been patched and can be turned into exploits immediately. National Security Agency officials testing Mythos have been impressed by the tool’s speed and efficiency in finding software vulnerabilities, according to a news report. Anthropic’s announcement of Mythos and the cybersecurity threat it poses garnered widespread media attention. Among the most widely reported were Mythos’ ability to identify a dormant 27-year-old security flaw in OpenBSD, a security-focused operating system, and a 16-year-old bug in FFmpeg, a video/audio processing tool. Some of these flaws allow unauthenticated users to gain control of the machines hosting these applications. Even more striking, the relatively inexperienced engineers running Mythos’ evaluations were able to use Mythos to complete attacks overnight, from finding vulnerabilities to exploiting them – something that can take human experts weeks to do. The model’s ability to chain multiple steps is what surprised Anthropic and organizations that tried it. In an evaluation by the AI Security Institute, Mythos was able to take over a simulated corporate network in three out of 10 tries, the first AI model to succeed at the task. These results are real. They also paint an incomplete picture in ways that matter. Where is the breakthrough? At first glance, Mythos’ breakthrough sounds novel and could signal a new class of cyber threats. However, a closer look suggests something different. The vulnerabilities Mythos found are not new in nature. They generally don’t belong to unknown security flaws, and in many cases they are variations of well-known and well-understood classes of software vulnerabilities. In cybersecurity, finding new instances of known types of flaws is not unusual. The most successful attacks rely on known, well-defined vulnerabilities that stay overlooked or unpatched. What concerned the researchers was not Mythos changing the nature of finding and exploiting vulnerabilities, but rather the intense scale and speed with which it was able to find and exploit those vulnerabilities. This is not a breakthrough per se but rather a result of decades of research in both cybersecurity and AI. In that sense, Mythos is the natural – and expected – result of powerful automation and AI integration because it follows the same fundamental procedures used in standard offensive cybersecurity practices. These include scanning for vulnerabilities, identifying patterns and testing exploitability. Mythos and similar emerging models make it possible to chain these steps together at a speed that is hard to fathom. So why were these vulnerabilities missed in the first place? It is crucial to understand that not all vulnerabilities are cost effective to fix, and not all vulnerabilities are a priority. Mythos did not discover a new kind of weakness – it exposed the limits of how cybersecurity practitioners search for them. New tech, age-old dynamic Mythos highlights an important fact about the reality of cybersecurity threats. System defenders are always at a disadvantage because they need to always succeed. Attackers, however, need to succeed only once to break the security of a system. This cat-and-mouse game will always be the same, and Mythos does not change that – it simply reinforces it. Mythos follows a familiar dynamic: A tool created to protect can also be used to attack and harm. “The same improvements that make the model substantially more effective at patching vulnerabilities also make it substantially more effective at exploiting them,” Anthropic officials wrote in a blog post about Mythos. What once may have required highly specialized skills can now be achieved with significantly less effort, which raises the most important question: Who will benefit first by using tools like Mythos – defenders or attackers? Mohammad Ahmad does not work for, consult, own shares in or receive funding from any company or organization that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.

Brock E.W. Turner / Axios: Enzo Health, whose AI tools help home health and hospice agencies automate tasks like patient intake and documentation review, raised a $20M Series A led by N47  —  Enzo Health, an AI workflow company for post-acute care, raised a $20 million Series A led by N47, CEO Zach Newman tells Axios Pro exclusively.

SAN FRANCISCO – The first global dataset showing the boundaries of agricultural fields was released in late April, after an 18-month campaign by geospatial experts in industry and academia. The […] The post Taylor Geospatial unveils global field dataset appeared first on SpaceNews.

AI tools speed up IoT development — but closer to the hardware, the same code that looks correct can silently break thousands of devices at once. The post How AI Tools Generate Technical Debt in IoT Systems — and What to Do About It appeared first on Towards Data Science.

Westend61/Getty Images Ethan Mollick, a Wharton professor and AI expert, said good taste is a key skill in the AI era. Mollick said on a podcast that standing out from AI-generated work matters. Tech leaders are debating whether "taste" is the next essential AI-era skill. Good taste is about well-cut clothes, chic interiors and, in the AI era, a potential strategy for standing out at work. Ethan Mollick, a Wharton professor told the April 30 episode of "The Education Equation with Jeremy Singer" that the technology is creating a "fundamental existential challenge" over how different kinds of intelligence are valued. As that shift happens, "having a sense of style actually matters," Mollick, the author of "Co-Intelligence: Living and Working with AI," said. In a world where most writing online is AI-generated and everything begins to sound the same, someone who can write with a "unique style ends up being more interesting," he said. As generative AI has begun to enter the workplace, anxieties have centered on technical fields such as computer science and finance, which are seen as highly exposed to AI job displacement. Some workers are finding reassurement in the idea that "human" skills, such as judgment, critical thinking, and communication, will remain harder to automate. But Mollick said on the podcast that even some of those traditional "durable skills" can now be partly outsourced to AI. "Those durable skills still seem durable, but they're also outsourceable to a degree they never were before," he said. "There's enough studies that say the AI gives pretty good critical thinking. It gives you pretty good answers on ethical situations that ethicists would agree with," he said, adding that they will only get better with time. Mollick's comments chime with what's become a topic of debate among some tech leaders in 2026 — whether taste is a key differentiator in the AI-powered workplace. In February, Greg Brockman, OpenAI's president, said that "taste is a new core skill," while Y Combinator cofounder Paul Graham posted on X: "When anyone can make anything, the big differentiator is what you choose to make." However, Nan Yu, head of product at Linear, wrote on X that "you probably don't have better taste than AI." He added: "There are plenty of other distinctly human things that we can contribute, but 'having better taste' isn't one of them." Read the original article on Business Insider

Choe Sang-Hun / New York Times: How AI tools such as SuperBrain and Naver's Talking Buddy are helping South Korea's elderly ease loneliness, detect emergencies, and slow cognitive decline  —  In the world's fastest aging society, artificial intelligence is being used to make care calls to older adults who live alone and to fight dementia.

Twenty years ago, this media brand didn't have a print edition to attract eyeballs and sponsors. Top-notch content and editorial talent did the heavy lifting.

Woven City is a privacy nightmare but could be helpful to an OEM desperate to be more.

Governance around Physical AI is becoming harder as autonomous AI systems move into robots, sensors, and industrial equipment. The issue is not only whether AI agents can complete tasks. It is how their actions are tested, monitored, and stopped when they interact with real-world systems. Industrial robotics already provides a large base for that discussion. […] The post Physical AI raises governance questions for autonomous systems appeared first on AI News.

Deborah Sophia / Reuters: Amazon debuts Supply Chain Services, which lets companies use its logistics network to move, store, and deliver everything from raw materials to final products  —  Amazon.com (AMZN.O) said on Monday it was rolling out “Amazon Supply Chain Services”, opening up its logistics network for other businesses to use.

Ionut Arghire / SecurityWeek: Instructure reported a data breach on April 30; ShinyHunters added Instructure to its victims list and claims it has 3.65TB of data from ~9K institutions  —  Hackers disrupted services and stole names, email addresses, student ID numbers, and user messages.

Finance chiefs need to think through three areas, says veteran CFO Amy Butte. Amy Butte This post originally appeared in the Business Insider Today newsletter. You can sign up for Business Insider's daily newsletter here. Is being a CFO in the AI era one of the toughest jobs going? With CEOs plowing big money into AI, it sometimes feels like that. We're starting to see success stories, but tangible results aren't yet guaranteed. And for CFOs holding the purse strings, it's like standing between a hungry dog and dinner. I decided to ask someone who's worn those shoes. Amy Butte is a four-time CFO, including at NYSE and, most recently, Navan. She said the CFO role has always been tough, but there are some "unique" challenges this time around. "It's always incumbent on a CFO to measure outcomes or measure impact and to translate the language of numbers into behavior or behavior into the language of numbers. But I think today we're dealing with rapid transformation, new usage of AI, new things to measure," Butte told me. "And so a CFO has to be willing to try new things — and let's remember, CFOs are not always risk takers." That doesn't mean CFOs need to reinvent the wheel. Butte said "old-school approaches" can work in this new environment. We talked through three key things for CFOs to be thinking about. Define success: Figure out what your investors are looking for to measure your growth. (Honestly, if you don't know that, you might have bigger problems.) Maybe it's revenue. Maybe it's pretax earnings. Maybe it's return on equity. Whatever the case, Butte said to make sure your AI bets are working toward that. Behind-the-scenes metrics: Now you can identify the key performance indicators that help you reach your ultimate goal (see above). Figure out a way to measure them. It could be comparing customer support interactions between AI and humans, or the percentage of code delivered successfully within a given timeframe. Don't be the one to stop innovation: "This is not the time for finance leaders to sit on the sidelines," Butte said. "It's important to take risks. It's important to try new things in an environment where change really can move the needle." None of this is set in stone. The measurements and definition of success can change, Butte said. Everyone just needs to be on board with the adjustments and make sure they're working together. She gave the example of someone who wants to get fit and lose weight. They might decide to track their steps, but they don't consider their diet, which might be derailing their plans. The result is a false sense of progress. That applies in the business world, where some companies are pushing hard for measurement tools like leaderboards and tokenmaxxing that don't ultimately lead to their bigger goals. "Just saying it is one thing. Measuring it — and measuring it with the right things, and sometimes it's multiple things — is critical. And to me, that's the responsibility of a CFO," Butte said. "Tokenmaxxing isn't going to solve for a bad product." Read the original article on Business Insider