Free-form protocols are essential for AI's task generalization but enable new threats like secret collusion and coordinated swarm attacks. Network effects can rapidly spread privacy breaches, disinformation, jailbreaks, and data poisoning, while multi-agent dispersion and stealth optimization help adversaries evade oversight--creating novel persistent threats at a systemic level. Despite their critical importance, these security challenges remain understudied, with research fragmented across disparate fields including AI security, multi-agent learning, complex systems, cybersecurity, game theory, distributed systems, and technical AI governance. We introduce multi-agent security, a new field dedicated to securing networks of decentralized AI agents against threats that emerge or amplify through their interactions--whether direct or indirect via shared environments--with each other, humans, and institutions, and characterise fundamental security-performance trade-offs. Our preliminary work (1) taxonomizes the threat landscape arising from interacting AI agents, (2) surveys security-performance tradeoffs in decentralized AI systems, and (3) proposes a unified research agenda addressing open challenges in designing secure agent systems and interaction environments. By identifying these gaps, we aim to guide research in this critical area to unlock the socioeconomic potential of large-scale agent deployment on the internet, foster public trust, and mitigate national security risks in critical infrastructure and defense contexts.Figure 1: Multi-agent threats demand multi-agent security: [Left] Two malicious AI agents (Mallory and Trudy) are interacting with a human user (Bob) through a shared message board seemingly innocuously to the overseer (magnifying glass).

Who you are, what you have experienced, and how you think inspires us to be innovative and bold. Loyal is an equal opportunity employer. We hire great people from a wide variety of backgrounds, not just because it's the right thing to do, but because it makes our company stronger. We welcome the unique contributions that you can bring in terms of your education, opinions, culture, ethnicity, race, ancestry, sex, gender identity and expression, national origin, citizenship, marital status, age, languages spoken, veteran status, color, religion, disability, sexual orientation, and beliefs. We consider qualified applicants regardless of criminal histories, consistent with legal requirements. Further, consistent with applicable federal and state law, Loyal provides reasonable accommodations when requested by qualified applicants or employees with disabilities, unless doing so would cause an undue hardship. Loyal's policy regarding requests for reasonable accommodation applies to all aspects of employment, including the application process.

A couple of weeks ago I created a ChatGPT chatbot on my tech help website. The bot was meant to help answer people's tech queries: it's ended up fielding questions way outside of its remit. In the fortnight it's been running, the chatbot has been asked how to build a magical potato, how to bring down totalitarian regimes and how to safely remove a remove USB stick from, shall we say, a delicate area. At least that query was tech related... Here's a round-up of the weirdest requests the AI bot has been forced to answer. Even though I explicitly instructed the AI bot not to answer questions that aren't related to tech, it sometimes can't help itself. Such as on this occasion, when it's dragged into a world of utter fantasy: AI: Sure, let's build a magical potato!

Vattenfall is one of Europe's largest producers and retailers of electricity and heat. Our main markets are Sweden, Germany, the Netherlands, Denmark, and the UK. The Vattenfall Group has approximately 20,000 employees. We have been electrifying industries, powering homes and transforming life through innovation for more than 100 years. We now want to make fossil free living possible within one generation and we are driving the transition to a sustainable energy system.

Phaidra is building the future of industrial automation. The world today is filled with static, monolithic infrastructure. Factories, power plants, buildings, etc. operate the same they've operated for decades -- because the controls programming is hard-coded. Thousands of lines of rules and heuristics that define how the machines interact with each other. The result of all this hard-coding is that facilities are frozen in time, unable to adapt to their environment while their performance slowly degrades.

Who you are, what you have experienced, and how you think inspires us to be innovative and bold. Loyal is an equal opportunity employer. We hire great people from a wide variety of backgrounds, not just because it's the right thing to do, but because it makes our company stronger. We welcome the unique contributions that you can bring in terms of your education, opinions, culture, ethnicity, race, ancestry, sex, gender identity and expression, national origin, citizenship, marital status, age, languages spoken, veteran status, color, religion, disability, sexual orientation, and beliefs. We consider qualified applicants regardless of criminal histories, consistent with legal requirements. Further, consistent with applicable federal and state law, Loyal provides reasonable accommodations when requested by qualified applicants or employees with disabilities, unless doing so would cause an undue hardship. Loyal's policy regarding requests for reasonable accommodation applies to all aspects of employment, including the application process.

Access to content on Oxford Academic is often provided through institutional subscriptions and purchases. Typically, access is provided across an institutional network to a range of IP addresses. This authentication occurs automatically, and it is not possible to sign out of an IP authenticated account. Choose this option to get remote access when outside your institution. Shibboleth / Open Athens technology is used to provide single sign-on between your institution's website and Oxford Academic.

Coral reefs are among Earth's most stunning and biodiverse ecosystems. Yet, due to human-induced climate change resulting in warmer oceans, we are seeing growing numbers of these living habitats dying. The urgency of the crisis facing coral reefs around the world was highlighted in a recent study that showed that 91% of Australia's Great Barrier Reef had experienced coral bleaching in the summer of 2021–22 due to heat stress from rising water temperatures. Determining reef health is key to gauging the extent of the problem and developing ways of intervening to save these ecosystems, and a new artificial intelligence (AI) tool has been developed to measure reef health using… sound. Research coming out of the UK is using AI to study the soundscape of Indonesian reefs to determine the health of the ecosystems.

This book was designed around major building blocks of the Python ecosystem that are useful to machine learning projects. There are a lot of things you could learn about Python, from language mechanics to the various libraries. Our goal is to take you straight to developing an intuition for the elements you can use in Python projects with laser-focused tutorials. We designed the tutorials to focus on how to get things done with Python. They give you the tools to both rapidly understand and apply each technique or operation. Each tutorial is designed to take you about one hour to read through and complete, excluding the extensions and further reading. You can choose to work through the lessons one per day, one per week, or at your own pace. I think momentum is critically important, and this book is intended to be read and used, not to sit idle. I would recommend picking a schedule and sticking to it.

A team of Swedish researchers has successfully used a printed artificial neuron to control the movement of a Venus flytrap plant, forecasting that it could aid the development of futuristic medical devices and brain-machine interfaces. Existing silicon-based circuits and devices, such as those used in current computers, are difficult to integrate with biological systems due to their complexity, poor biocompatibility and low energy efficiency. By contrast, the artificial neuron is based on organic electrochemical transistors, which more closely resemble electrical signalling systems found in biology. "The human brain is one of the most advanced computers ever made," says Simone Fabiano, an associate professor at Linköping University in Sweden and senior author on the study. "It has a massive amount of memory and is excellent at processing information and making decisions while consuming very little energy. On the contrary, man-made supercomputers are bulky and consume a lot of energy."