Today we’re launching Deep Agents deploy in beta. Deep Agents deploy is the fastest way to deploy a model agnostic, open source agent harness in a production ready way. Deep Agents deploy is built for an open world. It’s built on Deep Agents - an open

GoPro is cutting nearly 23% of its workforce amid declining revenue, signaling a cost-cutting shift… not AI-driven layoffs like others. The post GoPro Is Cutting Nearly a Quarter of Its Workforce (and AI Isn’t the Reason) appeared first on TechRepublic.

Darren Blanchard went a few seconds over his three minute time limit and found himself in handcuffs.

Discussion | Link

AI agents hold the promise of automatically moving data between systems and triggering decisions, but in some cases, they can act without a clear record of what, when, and why they undertook their tasks. That has the potential to create a governance problem, for which IT leaders are ultimately responsible. If an organisation can’t trace […] The post Agentic AI’s governance challenges under the EU AI Act in 2026 appeared first on AI News.

AI agents work best when they reflect the knowledge and judgment your team has built over time. Some of that is institutional knowledge that’s already documented and easy for an agent to use as-is. But most great organizations also rely on tacit knowledge that lives inside their employees’ minds.

Proteins rarely function in isolation as individual monomers. Most biological processes are governed by proteins interacting with other proteins, forming... Proteins rarely function in isolation as individual monomers. Most biological processes are governed by proteins interacting with other proteins, forming protein complexes whose structures are described in the hierarchy of protein structure as the quaternary representation. This represents one level of complexity up from tertiary representations, the 3D structure of monomers… Source

Gorodenkoff/Shutterstock Not long ago, if you needed a speech polished, a document translated or a logo designed, you would probably have hired a freelancer online. Millions of people did exactly that. They went to platforms such as Fiverr and Upwork and paid someone (maybe on the other side of the world) to do the job. In 2023, online gig workers were estimated to number between 154 million and 435 million globally. As such, they could represent as much as 12.5% of the global labour force. Today, however, many people do something else. They open ChatGPT. Generative AI now acts as a copy editor, translator, illustrator and research assistant in one. It can summarise a report in seconds, write social media posts, create a presentation or produce a simple logo at virtually no cost. What, then, has happened to the freelancers who used to do this work? AI has long been discussed as a threat to jobs and livelihoods. But what’s the reality? In this new series, we explore the impact it is already having on different occupations – and how people really feel about their AI assistants. Some freelancers are struggling. But perhaps surprisingly, others are doing better than ever. Demand and wages have fallen for some kinds of online freelance work. Translation, basic copywriting and simple graphic design have been hardest hit. According to one study, demand for freelance writers was found to have fallen by up to 30% after the release of generative AI tools. Other research suggests that freelancers who are highly exposed to AI saw earnings fall by as much as 14%. Yet there is also evidence that many freelancers are thriving. Freelancer platform Upwork reports that higher-value contracts – those worth more than US$1,000 (£745) – increased across various disciplines after the arrival of generative AI. Freelancers using AI-related skills earn around 40% more than comparable freelancers who do not. How can both of these things be true? The answer becomes clearer when you stop thinking about “freelancers” as one group and instead look at the tasks and skills they perform. Some kinds of freelance work are highly commodified. They consist of narrowly defined, repetitive tasks that can be clearly described and easily compared. This could be things like translating a document, summarising a report, drafting a press release or designing a basic logo. These tasks are exactly what generative AI is good at. They rely on patterns, templates and predictable instructions. The more closely a freelancer’s work resembles the tasks that AI can perform, the more likely it is to come under pressure. But other freelancers do not sell a single narrow skill. They sell a more complex bundle of expertise. A legal translator does not merely convert words from one language to another. They understand legal terminology, cultural nuance and the risks of getting a phrase wrong. Similarly, a branding consultant combines design with market research and consumer psychology. A software developer may use AI to generate code, but still needs to understand the client’s business problem to decide which solution actually works. These workers can use AI to automate the repetitive parts of their jobs while concentrating on the aspects that clients still value most: expertise, judgment and trust. Online today, in the office tomorrow This matters far beyond online freelancing platforms. Online labour markets often act as an early warning system for the wider economy. This work is more transactional and less protected by the institutions that shape conventional employment (things like long-term contracts, internal promotion ladders and unions). Because tasks are posted, bought and completed on the open market, technological change shows up there more quickly than in ordinary workplaces. What happens on Fiverr or Upwork today may happen in offices tomorrow. This is already becoming visible in law firms, consultancy companies and marketing agencies. Many junior employees spend much of their time summarising documents, preparing presentations, drafting reports or conducting basic research. These are precisely the kinds of tasks that AI can perform. Recent evidence from the US labour market suggests that younger and less-experienced workers are already bearing the brunt of AI-related disruption. Senior workers, by contrast, tend to do more complex work, combining technical knowledge with experience and human interaction. Legal translation is more than converting words from one language to another. CrizzyStudio/Shutterstock The response should not be to compete with AI at the things AI already does well. Instead, workers need help building deeper forms of expertise and combining skills in ways that are harder to automate. This is in the interest of workers, but also of the platforms themselves. Fiverr, Upwork and others promise clients efficient and high-quality work. If routine tasks are increasingly automated away, they will depend more heavily on workers who can offer something more than a standardised service. That means platforms should actively provide skill-building courses, training resources and guidance on how to use AI productively. They could also offer micro-credentials that certify newly acquired expertise. These credentials have been found to help workers enter online labour markets and increase their earnings. The challenge, then, is not to stop people from using AI. It is to ensure that workers are not trapped in forms of work that are so narrow, standardised and commodified that they can easily be automated away. The future of online (and onsite) work may depend less on whether we use AI than on whether our jobs can be reduced to something an AI can easily imitate. Fabian Stephany does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.

The product comes amid a flurry of releases from the generative AI vendor, which is locked in a tight battle with rival OpenAI.

London-based Qoro Quantum has secured $750,000 in a pre-seed funding round to develop softwareinfrastructure for hybrid quantum-classical computing. The round includesbacking from Ada Ventures, Supera...

The case was the first time authorities charged people for alleged “Antifa” activities after President Trump designated the umbrella term a terrorist organization.

AI tax prep automation startup Juno, founded to address the opportunities — and risks — that come with advances in AI, and built for underserved SMB accounting firms, says it has received a $12 million seed investment.

Robotic cloud laboratories powered by AI can carry out experiments remotely and cut costs. J Studios/DigitalVision via Getty Images Artificial intelligence is rapidly learning to autonomously design and run biological experiments, but the systems intended to govern those capabilities are struggling to keep pace. AI company OpenAI and biotech company Ginkgo Bioworks announced in February 2026 that OpenAI’s flagship model GPT-5 had autonomously designed and run 36,000 biological experiments. It did this through a robotic cloud laboratory, a facility where automated equipment controlled remotely by computers carries out experiments. The AI model proposed study designs, and robots carried them out and fed the data back to the model for the next round. Humans set the goal, and the machines did much of the work in the lab, cutting the cost of producing a desired protein by 40%. This is programmable biology: designing biological components on a computer and building them in the physical world, with AI closing the loop. For decades, biology mostly moved from observation toward understanding. Scientists sequenced the genomes of organisms to catalog all of their DNA, learning how genes encode the proteins that carry out life’s functions. The invention of tools like CRISPR then allowed scientists to edit that DNA for specific purposes, such as disabling a gene linked to disease. AI is now accelerating a third phase, where computers can both design biological systems and rapidly test them. The process looks less like traditional benchwork in a lab and more like engineering: design, build, test, learn and repeat. Where a traditional experiment might test a single hypothesis, AI-driven programmable biology explores thousands of design variations in parallel, iterating the way an engineer refines a prototype. As a data scientist who studies genomics and biosecurity, I research how AI is reshaping biological research and what safeguards that demands. Current safety measures and regulations have not kept pace with these capabilities, and the gap between what AI can do in biology and what governance systems are prepared to handle is growing. What AI makes possible The clearest example of how researchers are using AI to automate research is AI-accelerated protein design. Proteins are the molecular machines that carry out most functions in living cells. Designing new ones has traditionally required years of trial and error because even small changes to a protein’s sequence can alter its shape and function in unpredictable ways. Protein language models, which are AI systems trained on millions of natural protein sequences, can quickly predict how mutations will change a protein’s behavior or design new proteins. These AI models are designing potential new drugs and speeding vaccine development. Paired with automated labs, these models create tight loops of experimentation and revision, testing thousands of variations in days rather than the months or years a human team would need. Faster protein engineering could mean faster responses to emerging infections and cheaper drugs. The dual-use problem Researchers have raised concerns that these same AI tools could be misused, a challenge known as the dual-use problem: Technologies developed for beneficial purposes can also be repurposed to cause harm. For example, researchers have found that AI models integrated with automated labs can optimize how well a virus spreads, even without specialized training. Scientists have developed a risk-scoring tool to evaluate how AI could modify a virus’s capabilities, such as altering which species it infects or helping it evade the immune system. Current AI models are able to walk users through the technical steps of recovering live viruses from synthetic DNA. Researchers have determined that AI could lower barriers at multiple stages in the process of developing a bioweapon, and that current oversight does not adequately address this risk. Robots can carry out human- or AI-designed studies in the lab. Du Yu/Xinhua via Getty Images Risk from bio AI Experienced scientists are already using AI to plan and design biological experiments. The question of whether AI can help people with limited biology training carry out dangerous lab work is the subject of active research. Two recent studies have reached different conclusions. A study by AI company Scale AI and biosecurity nonprofit SecureBio found that when people with limited biology experience were given access to large language models, which is the type of AI behind tools like ChatGPT, they were able to complete biosecurity-related tasks, such as troubleshooting complex virology lab protocols with four times greater accuracy. In some areas, these novices outperformed trained experts. Around 90% of these novices reported little difficulty getting the models to provide risky biological information, such as detailed instructions on working with dangerous pathogens, despite built-in safety filters meant to block such outputs. In contrast, a study led by Active Site, a research nonprofit that studies the use of AI in synthetic biology, found that AI help did not lead to significant differences in the ability of novices to complete the complex workflow to produce a virus in a biosafety laboratory. However, the AI-assisted group succeeded more often on most tasks and finished some steps faster, most notably on growing cells in the lab. Hands-on work in the lab has traditionally been a bottleneck to translating designs into results. Even a brilliant study plan still depends on skilled human hands to carry out. That may not last, as cloud laboratories and robotic automation become cheaper and more accessible, allowing researchers to send AI-generated experimental designs to remote facilities for execution. Responding to AI-driven biological risks AI systems are now able to run experiments autonomously and at scale, but existing regulations were not designed for this. Rules governing biological research do not account for AI-driven automation, and rules governing AI do not specifically address its use in biology. In the U.S., the Biden administration had issued a 2023 executive order on AI security that included biosecurity provisions, but the Trump administration revoked it. Screening the synthetic DNA that commercial providers make to ensure it cannot be misused to make pathogens or toxins remains mostly voluntary. A bipartisan bill introduced in 2026 to mandate DNA screening does not yet address AI-designed sequences that evade current detection methods. The 1975 Biological Weapons Convention, an international treaty prohibiting the production and use of bioweapons, contains no provisions for AI. The U.K. AI Security Institute and the U.S. National Security Commission on Emerging Biotechnology have both called for coordinated government action. The safety evaluations that AI labs run before releasing new models are often opaque and unsuited to capture real-world risk. Researchers have estimated that even modest improvements in an AI model’s ability to help plan pathogen-related experiments could translate to thousands of additional deaths from bioterrorism per year. Timelines for when these capabilities cross critical thresholds remain unclear. The Nuclear Threat Initiative has proposed a managed access framework for biological AI tools, matching who can use a given tool to the risk level of the model rather than blanket restrictions. The RAND Center on AI, Security and Technology outlined a set of actions researchers could take to improve biosecurity, including improved DNA synthesis screening and model evaluations before release. Researchers have also argued that biological data itself needs governance, especially genomic data that could train models with dangerous capabilities. Some AI companies have started voluntarily imposing their own safety measures. Anthropic activated its highest safety tier when it released its most advanced model in mid-2025. At the same moment, OpenAI updated its Preparedness Framework, revising the thresholds for how much biological risk a model can pose before additional safeguards are required. But these are voluntary, company-specific steps. Anthropic’s CEO, Dario Amodei, wrote that the pace of AI development may soon outrun any single company’s ability to assess the risk of a given model. When used in a well-controlled setting, AI can help scientists quickly reach their research goals. What happens when the same capabilities operate outside those controls is a question that policy has not yet answered. Overreact, and talent and investment may move elsewhere while the technology continues advancing anyway. Underreact, and the risks of that technology could be exploited to cause real harm. I am applying for grants in this area.

Discussion | Link

Anthropic’s most capable AI model has already found thousands of AI cybersecurity vulnerabilities across every major operating system and web browser. The company’s response was not to release it, but to quietly hand it to the organisations responsible for keeping the internet running. That model is Claude Mythos Preview, and the initiative is called Project Glasswing. […] The post Anthropic keeps new AI model private after it finds thousands of external vulnerabilities appeared first on AI News.

Discussion | Link

The Nigerian village of Mmiata Anam in Anambra state, completely submerged by floods. Chinedu Chime/Shutterstock A message appears online during heavy flooding: “This rain no be small o, everywhere don red.” Someone unfamiliar with the phrasing might hesitate. But for people in Nigeria, this message is immediate and clear: the flooding is severe and worsening. Moments like this happen all the time on digital platforms. People don’t write in perfect, standard English sentences. They share warnings and reactions on platforms like X, WhatsApp and Facebook using the language of everyday life. This means sometimes mixing English with local expressions, slang and expressive language shaped by their communities. Artificial intelligence systems can understand language and tackle a wide range of problems. Governments and organisations are increasingly using AI to scan social media, summarise public conversations, and even respond to environmental and climate issues. But many of these tools struggle to make sense of the way people actually communicate. Local expressions and slang can confuse AI, so important messages are sometimes misunderstood or missed entirely. When people talk about language barriers, they often mean translation between different languages. But the problem is more subtle. Around the world, people mix languages and local expressions online, a phenomenon that linguists call “code switching”. Climate journalism has increasingly moved online, but there are fewer climate reporters in the developing world. This limits the depth and availability of information for a huge proportion of the global population, and shapes how climate issues are discussed and understood across different regions. For instance, a UK social media post might raise an environmental concern using expressions like: “Are roads flooding already? Chuffed to know the council taking the piss.” Most AI tools can pick up the sarcasm and frustration aimed at local authorities. In a country such as Nigeria, people may describe unfolding concerns differently: “Abeg is it October wey rain dey fall like this, but you say the climate no change?” or “River don near our house o! Abeg help, e fit spoil everything!” Here, slang and Pidgin express immediate danger and an urgent call for help. Yet AI models often diminish this to casual commentary, entirely missing the urgency and emotion that is being conveyed. This matters because most AI systems are taught on large western-centric text, mainly from North America and Europe. ChatGPT, for example, is instructed on huge amounts of internet text. It doesn’t have beliefs, feelings or awareness. Instead, it generates responses based on patterns it has seen online. AI reflects the dominant culture in its training data, so carries a “cultural fingerprint”. It imitates normal ways of expressing ideas from the societies that produced the texts it has learned from. AI models trained on predominantly English-language texts show a hidden bias that favour western cultural values, particularly when asked in English. Flash floods in Wawa, a communtiy in south-western Nigeria followed heavy torrential rainfall in 2019. Oluwafemi Dawodu/Shutterstock One major reason AI can produce biased outcomes is that it reflects the societal inequalities including differences in race, gender and region that show up in the data it learns from. So, underrepresented voices from communities in developing countries with non-Anglocentric varieties of English are often diminished or ignored. This bias can have real consequences. In climate crises like floods, heatwaves or other extreme weather, misinterpreted messages could put property and lives at risk. AI systems that rely on past patterns are easy to interpret when language fits expected standards, but posts that don’t conform with the presence of local slang or urgency cues can be misinterpreted. Improving climate disaster responses Solving this problem involves designing systems that actually reflect the way people communicate. AI systems need to be trained to understand regional expressions and recognise that meaning often depends on cultural context, not just literal words. AI should be tested on real online posts, not formal western-centric English, to capture urgency and local references. Automated systems can process huge volumes of information, but human judgment must remain in the loop – especially when people’s safety is at stake. AI tools can help communities respond to floods, heatwaves and other climate emergencies – but only once trained to interpret the nuance of everyday language, so that warnings and calls for help get through. Ifeoluwa Wuraola receives PHD funding from Centre of Excellence for Data Science, Artificial Intelligence and Modelling (DAIM). Daniel Marciniak and Nina Dethlefs do not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and have disclosed no relevant affiliations beyond their academic appointment.

The conflict in Iran – but also the war in Ukraine – show not only that AI is radically changing the economics of war (which may be good news), but also that we may be heading towards some kind of “Chernobyl moment”. We may soon experience a disaster that will force us to belatedly realise we should have drawn up some shared rules to govern a technological development that we ourselves triggered. Even Dario Amodei, the founder of AI company Anthropic, who seems passionate about taking action to prevent Armageddon, acknowledges that he doesn’t have the answer we desperately need. One of the most interesting attempts to regulate the use of artificial intelligence may have been the one drafted during the second world war by a PhD student at Columbia University who was then temporarily employed by the US Navy. His name was Isaac Asimov, and in his early short story Runaround (1941), he postulated three laws that are still surprisingly inspiring for anyone thinking about how to solve the intellectual and political problem that is AI in warfare. Unlike recent attempts by the OECD (Organisation for Economic Co-operation and Development) and the EU to draw up regulations, Asimov’s laws are admirably concise. They state that a robot (what we now call an “artificially intelligent agent”) shall never harm a human being (or allow harm to happen through inaction. It shall always obey the orders given by humans unless they conflict with the first prohibition. And it will always protect its existence unless this conflicts with the first and second provisions. In his story, Asimov himself shows how the three laws can create internal contradictions, leading to paralysis. And yet, Asimov’s three principles can still be useful as a starting point for the strategy we now need. Anthropic takes a ‘stance’ The biggest merit of the note Dario Amodei wrote recently on the perils of a technology which is still in its adoloscence is the acknowledgement that Anthropic, the firm that Amodei founded, is using its own large language model (called Claude) to develop further versions of itself. Artificial intelligence is generating even more intelligent robots and this brings us near to that “singularity” first theorised by the great mathematician John von Neumann – the moment when artificial intelligence surpasses human intelligence and renders us irrelevant. If the technology is an adolescent, it is growing very fast and will soon be out of the control of its creator. Amodei does not, however, appear to have a concrete proposal on how to manage this problem. He has said that Anthropic’s contracts with the US Department of War should never include the use of the company’s models for empowering either “mass domestic surveillance” or “fully autonomous weapons”. It is a request that has brought Anthropic into a bitter dispute with the US government. And yet it seems a rather narrow response that covers just one dimension of a much wider problem. Amodei focuses predominantly on the safety of US citizens when it is people elsewhere in the world who are currently most affected by the use of autonomous weapons. We need a bolder vision – and Asimov’s intuitions may help. New rules One approach would be to ask all developers of AI models to introduce in their foundational codes three simple and bold commands along the lines of: “You will never kill a human being (unless for self-defence)”; “you will always try to work for the betterment of mankind (unless such a provision entails the violation of the first command)”; “when you doubt that your actions may violate the first or the second commands, you will choose inaction and ask what to do”. Most likely, this initiative will have to come from a group of countries following a pattern similar to the treaties of non-proliferation of nuclear weapons. And it would be good to have a debate on some new ideas before we are forced to do so by some AI-empowered nuclear unintended consequence. Like all other attempts to regulate a future that we still cannot even envisage, the three commands will have some drawbacks. A robot may have refused to kill Iran’s former leader Ali Khamenei, but that may be a price worth paying if it means we can avoid setting a precedent for other discretionary and dangerous interpretations. Robots may not always be successful at identifying human beings (as Asimov himself acknowledged in later writing) and yet this may well be one of those intellectually fascinating problems that models born to make sense of human language will solve. More importantly, it will take not only information but a lot of wisdom to understand what is good for humankind. Robots may end up sitting frequently idle waiting for instructions. And yet efficiency is not a religion we have to follow when the challenge is about the survival of our species. Making sense of what increasingly appears to be one of the greatest technological revolutions of all time requires careful thought and forward planning. Francesco Grillo is Director at Vision, the think tank.

Startups in Latin America raised a combined $1.03 billion across seed- and growth-stage deals in the three-month period ending March 31 — up year over year, but down from Q4 2025.

In Seattle, activists inside tech giants are leaning into grassroots fundraising and peer support as petitions and protests draw little response from executives.