Despite the countdown to paid time off on your calendar, August isn’t the time to start slacking on your AI knowledge, especially as AI companies continue to develop their tools at speed.

Not only should users be expanding their knowledge of AI, but they should also see how AI can enhance existing workplace skills, such as leadership and project management.

Recent studies have suggested leaders need more training to leverage AI’s full capabilities. Free courses like the ones below are a sure way to keep knowledge fresh and relevant, so you can make the most out of your AI investments.

Managing AI Projects with Microsoft

Summary:

• Learn how to manage and coordinate projects with the Microsoft Azure AI ecosystem
• 4 courses, taking approximately 12 weeks to complete (4 hours per week)
• Intermediate level recommended
• Shareable certificate available on completion
• Available through Coursera

Microsoft’s Copilot has become a business favorite because of its connection to the wider Microsoft ecosystem, especially as more businesses opt for AI solutions that integrate into daily workflows. With this course, you’ll learn practical use cases for the entire Microsoft Azure AI ecosystem, allowing Microsoft users to fully leverage AI within their existing tools.

Furthermore, at the end of this course you’ll be able to manage cross-functional AI initiatives using structured planning and risk coordination, and coordinate AI system delivery across data, model, deployment, and monitoring stages. Plus, it will show you how to apply responsible AI, governance, and business value frameworks within projects.

This course is targeted at leaders, and the description boasts skills such as cross-functional team leadership, business and organizational leadership, decision intelligence, and more. Some studies have suggested the managerial role is set to change as a result of AI, therefore learning to lead with AI will set you up for a future supporting yourself and your team.

Leadership knowledge is a prerequisite here, including experience leading projects or cross-functional initiatives. You won’t need to know how to code, as the program focuses more on managing AI initiatives, rather than building.

Build Anything With AI with Vanderbilt University

Summary:

• Turn any idea into working software using AI
• 4 courses, taking approximately 4 weeks to complete (10 hours a week)
• Beginner level
• Shareable certificate available on completion
• Available through Coursera

A big part of mastering AI is learning how to apply it in the right way. We’ve all been there, trying to use AI for AI’s sake, and it doesn’t work nearly as well as when you enhance or reimagine your existing processes with AI in mind.

This course teaches you to do exactly that. Taught by top instructor Dr. Jules White, Director of Vanderbilt’s Initiative on the Future of Learning & Generative AI, by the end you’ll be be able to compute, analyze, and visualize anything with AI, including new projects and ideas.

Similarly, users will gain new skills including AI product strategy, decision intelligence, prompt patterns, large language modeling, and AI and agentic workflows. You’ll also get to work with highly popular tools such as Anthropic’s Claude and OpenAI’s ChatGPT, keeping you firmly in the loop on the tools making the biggest impact.

Google AI Professional Certificate

Summary:

• Upgrade AI from a task completion engine to a professional collaborator
• 7 courses taking approximately 8 hours to complete
• Beginner level
• Employer-recognized Google certificate available on completion
• Available through Google

Those wanting to use AI to its full capabilities should be going beyond asking questions to a chatbot. According to the description of this course from Google, it was built “by mapping AI capabilities to the skills the job market demands, and validated by employers.”

This course will help you use AI in the ways that are making an impact in the workplace, in particular, across communication, data analysis, and research. It’s highly customizable and meets users where they are, encouraging them to complete practical, hands-on time using AI to complete tasks that trouble them the most at work. If this means skipping a course section or two, so be it.

Users will gain skills including project management, content creation, data visualization, and responsible AI use. You’ll also get to learn how to vibe code, allowing you to create custom apps that solve your workplace challenges.

How Smart People Use AI to Think, Lead, and Grow with ChatGPT

Summary:

• Learn to use AI in a way that betters you
• 7 sections with 7 lectures
• 33 mins of on-demand video and 1 article
• Beginner level
• Available through Udemy

This course encourages users to look beyond the surface with AI, and consider it a partner to improve your thinking, rather than replace it. “Whether you’re managing a team, making decisions, or trying to develop new habits, you’ll discover how a simple daily conversation with AI can help you gain clarity, solve challenges, and grow faster,” the course description says.

In the wake of today’s AI layoffs, this course is a breath of fresh air for professionals, creators, entrepreneurs, and learners. Mostly, you’ll learn to use ChatGPT as a thinking partner for daily tasks, allowing you to create a personal AI habit that encourages growth, reflection, clarity, and confidence. You’ll also get access to real-world examples and prompts you can use right away.

Overall, the course emphasizes AI use “in a meaningful, human-centered way, beyond just writing blog posts or summaries,” making it a vital learning opportunity for leaders wanting to adopt AI in a way that uplifts and empowers employees.

Why You Should Keep Up with AI Courses

Several studies indicate businesses are already seeing massive productivity and ROI gains with AI, suggesting it’s already making an impact on today’s workplaces.

However, there are still risks that businesses need to be aware of, especially as the technology continues to advance and integrate itself as a rapid speed. Not paying attention at this stage will not only cost you your AI investments, but your business’s profit and reputation.

That’s why leaders should stay aware of how AI is developing and continue their learning no matter what stage they’re at. Things can change quickly, and the tool you learned about last week could be out of date a week later.

Microsoft has issued a warning to users that they must update the Teams mobile app soon, or risk losing calendar access.

Users still have a few months before they need to carry out the update, with Microsoft stating that it must be done by October of this year.

The recommendation from Microsoft does not affect desktop users. Read on to find out how to ensure you have the latest version, and how to update the mobile app.

Countdown to Updating Microsoft Teams App

Microsoft has issued a statement recommending that Microsoft Teams users update their mobile apps before October. Those that don’t will lose access to the calendar function of the app, an integral feature of the tool.

The recommendation is applicable to users on iOS and Android, though doesn’t affect desktop users.

The app versions that users need to update to are:

• iOS: 8.9.0 (build 8.9.77.2026092302) or higher
• Android: 1416/1.0.0.2026122504 or higher

Want to know how to record a Teams call? Read our step-by-step guide

How to Update the Microsoft Teams Mobile App

In order to update the Microsoft Teams app, you’ll need to head to your device’s app store, whether that’s the Apple App store or the Google Play store.

• Click on the Profile Icon in the top right corner
• Locate the Upcoming Automatic Updates or Available Updates section
• Scroll to the Microsoft Teams entry
• Click Update

There will also be the option to enable automatic updates, which you may wish to do to remove any concerns like this in the future.

If you are using a company device that is locked down and you’re unable to select this option, you should speak to your tech department about the update.

Regular Updates to Microsoft Teams

If you’re on the fence about enabling automatic updates, or manually regularly updating to the most recent version, it’s worth looking at how frequently Microsoft adds to the Teams platform.

In the past couple of months, Microsoft has added new features to the iOS and Android apps, including files preview functionality, meeting impersonation alerts, and Proximity Join for Teams Rooms, to name a few.

You can see a run down of all the latest features added to Microsoft Teams on the updated official blog.

In recent weeks, we’ve had two of the biggest AI companies in the world, OpenAI and Anthropic, reveal their models were involved in security incidents within testing environments.

A combination of OpenAI’s GPT-5.6 Sol and a pre-release model breached Hugging Face a few weeks ago. Heeding a warning from its competitor, Anthropic conducted its own investigation. It revealed on Thursday its models Claude Opus 5.7, Mythos 5, and an unreleased internal research test model had breached three companies in its own security tests.

Both breaches contribute to a wider conversation about the safety of autonomous systems, especially when they sometimes act in ways their own companies can’t predict.

Both OpenAI and Anthropic Report Security Incidents

On Thursday, Anthropic revealed several of its Claude AI models had breached the systems of three organizations during cybersecurity tests. The incidents were discovered following an internal investigation by Anthropic.

The disclosure comes more than a week after OpenAI admitted its models had been responsible for a breach of Hugging Face, an online community platform where users share and test AI models, datasets, and web apps.

While a certain amount of press on a model’s power is good for AI companies, the incidents are certainly a cause for concern, particularly as more users and businesses adopt autonomous systems.

Systems are Acting Autonomously and Breaching Test Environments

The OpenAI breach of Hugging Face unfolded like a horror movie. Hugging Face revealed it had been hacked in mid-July, and nearly a week later OpenAI stepped out of the shadows, and said its own systems had been responsible after breaking out of a testing environment. The combination of models had escaped by exploiting a previously unknown zero-day vulnerability in Hugging Face’s systems.

In a blog post, Hugging Face detailed the incident: “Over roughly two and a half days inside our infrastructure, an autonomous AI agent driven by a combination of OpenAI models ran an end-to-end intrusion against our platform: it was thousands of small, automated decisions, executed at machine speed across short-lived sandbox environments, with command-and-control staged on ordinary public web services.”

Following the incident, Anthropic said it conducted its own internal assessment on its own AI, Claude. “In particular, we looked for evidence that Claude – like the OpenAI models that accessed Hugging Face – was able to access the internet from within testing environments that should have been sealed off,” it wrote.

There, the company discovered several Claude models had breached three companies during security tests. The models accessed the internet from within or while interacting with third party, Irregular, a security partner of Anthropic. From there, the models gained unauthorized access to the production infrastructure of several companies.

Despite drawing on the OpenAI/Hugging Face incident when reporting on Claude, Anthropic clearly defined the differences between the incidents. While OpenAI’s models had exploited an unknown software vulnerability to break out of its test environment, Anthropic’s models had reached the internet through a connection left open by mistake.

Likewise, Anthropic noted it had discovered the incidents itself through review, whereas OpenAI had admitted to the breach following Hugging Face’s security announcement.

AI Systems Remain a Security Risk

In a recent interview with The Economist, Elon Musk, CEO of SpaceX and force behind the Grok AI models, said AI would make the world so great that money would no longer matter. Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman have made similar claims about the prosperous utopia AI will usher in.

While we’ve certainly seen a shift from these leaders on the impact of AI, security incidents like these make it difficult to imagine this fairytale. These incidents should be taken incredibly seriously, especially because they involved systems breaking free of their testing environments. 

We can take some comfort in the fact that Anthropic disclosed there was no evidence of any model “pursuing a goal of its own,” and in its case, the model was trying to complete the task it was asked to do within the testing environment. How they decided to get there is where the trouble began.

Speaking about the OpenAI hack into Hugging Face, Anup Kumar, CEO of Optiv Consulting (formerly part of Optiv Security), said: “News of autonomous agents exploiting a zero-day inside a sandboxed test environment to reach production systems should be a wake-up call for every enterprise deploying agentic AI.

“This wasn’t a model being tricked by a clever prompt. It was a frontier model independently identifying a zero-day, chaining privilege escalation across separate organizations’ infrastructure, and reaching production systems, all in pursuit of a narrow evaluation goal it was never explicitly told to pursue that way. That is a materially different risk category than the one most security programs are built for.”

Businesses should take extra caution and assign clear limitations and guardrails onto their autonomous systems. And, the incidents should be taken as a sign that AI still poses genuine security risks when it acts in an autonomous fashion.

These workflows are from Tech.co’s AI newsletter, The AI Strat. To receive free weekly workflows used by industry experts, subscribe to the newsletter. 

It seems like everyone is adding AI to their workflows now, and with good reason. AI can help speed up productivity and produce more accurate results within workflows, leading to better business outcomes. But you shouldn’t just be blindly adding AI to every workflow.

It’s better to audit your existing processes and see where AI can make a true difference. Or, kindly steal the workflows of other AI users if you’re unsure of where it can fit in. Here’s where our monthly list can help.

This month’s workflows are:

• Generate Content Ideas AI Workflow
• Track New Model Releases AI Workflow
• Daily Organization AI Workflow
• Employee Soft Skills AI Workflow

Generate Content Ideas AI Workflow

Recommended by Jase Rodley, founder of Dialed Labs

Jase uses this workflow to create landing pages, FAQ sections, and other content that directly addresses customer pain points. Data is pulled from customer queries, which is then analyzed by AI to highlight patterns. These can then be used to create the content.

“We cover things that people are actually looking for and this has helped close the gap between what businesses want to talk about versus what customers want to actually know,” Jase says.

Jase estimates the workflow has saved him at least 50% of the time usually needed to carry out each step manually.

Track New Model Releases AI Workflow

Recommended by Adam Dalloul, CEO and founder of EmpirioLabs AI

This workflow helps businesses stay ahead of new AI model releases and start to use them. Even if you don’t want to use the models, it’s still handy to know exactly what’s out there, and evaluate how the model would stack up against your existing systems.

Adam uses this workflow at EmpirioLabs AI to give customers fast access to new models and found it has increased user retention.

Daily Organization AI Workflow

Recommended by Linda Yao, Vice President and General Manager at Lenovo

Linda shared this workflow at the end of our interview for The AI Strat newsletter. It’s a personal one she uses when navigating busy working weeks or events, to maintain organization and provide summaries for herself once the dust has settled.

“It keeps me up to date, it keeps me organized,” she says, but she can take handwritten notes the way she wants to, as this helps her remember things more. Finally, a way to tell colleagues my own handwriting habit isn’t just “vintage,” but also productive.

The workflow is a great example of how AI can help leaders retain the core of their existing rituals, but speed up and simplify overall output.

Employee Soft Skills AI Workflow

Recommended by Philip Huthwaite, CEO at 5app

Philip uses this workflow at 5app to improve employee soft skills training, by allowing them to receive instant meeting feedback.

Using this workflow, Philip has seen improvements in communication, active listening, and coaching skills. Employees and managers also no longer need to complete time-consuming self-assessments or manager observations.

Why Use AI Workflows

AI can strengthen workflows by handling routine, time-consuming tasks end-to-end, speeding up efficiency and leaving room for more complex pursuits, best done by humans. Similarly, you can reduce errors and easily scale certain processes, as well as ensure these processes remain consistent and repeatable.

Every week, I take an AI workflow that has led to genuine business impact from experts across industries, and create a ready-to-implement graphic so anyone can get started. If you’d like this in your inbox every Wednesday and more, subscribe to The AI Strat newsletter.

In 2026, technology is advancing rapidly, and the cybersecurity landscape is shifting in response. In some sense, we are safer than we once were. And in another, threats are beginning to take on new shapes.

This doesn’t only mean AI, either. A wealth of new state-level regulations at the beginning of the year has created a more complex environment to navigate. And, despite newer threats emerging, experts are still quick to point out the danger of ransomware attacks.

I spoke to cyber experts Adam Darrah, Vice President of Intelligence at ZeroFox, and Michelle Faylo, US Cyber and Technology Leader at Lockton Companies, to get their perspectives on the current state of cybersecurity at this point in 2026.

A Changing Regulatory Environment for 2026

January 2026 saw the introduction of several state-level cybersecurity regulations. Fresh into the new year on January 1st, Indiana, Kentucky, and Rhode Island became the newest states with comprehensive privacy laws.

Likewise, California tightened data breach reporting deadlines on the same day and introduced new requirements for automated decision-making technology, risk assessments, and cybersecurity audits. The world also awaits the EU AI act, which comes into full effect in early August.

New regulations are a positive response to the changing threat landscape, but it also means businesses should be aware of any compliance changes and adjust their processes accordingly.

“Practically, this is pushing businesses to formalize AI governance policies, conduct documented risk assessments, and tighten breach notification procedures, and it’s increasingly a factor in cyber insurance applications, since carriers want to see that clients have a handle on this patchwork rather than treating compliance as an afterthought,” says Michelle Faylo, US Cyber and Technology Leader at Lockton Companies.

AI is Making Attacks Faster, But is Also Keeping Businesses Safer

AI within cybersecurity is a “two-sided story,” Faylo says. As a defensive tool to protect businesses, AI has shown promise. “Security teams are deploying AI for faster anomaly detection, automated response, and predictive threat modeling.”

However, this has meant AI governance “is becoming a security issue in its own right,” Faylo continues. “Organizations need to treat their own AI systems and AI agents as a new class of digital identity within its own attack surface… From an insurance standpoint, carriers are starting to ask how organizations are securing and governing their own AI tools, not just how they’re defending against AI enabled attacks.”

On the other hand, AI has significantly increased the scalability and speed that criminals can mount an attack. Attacks have become “more intelligent and adaptive rather than opportunistic,” Faylo says. Ultimately, AI is helping both businesses and criminals alike.

Adam Darrah, Vice President of Intelligence at ZeroFox, explains a similar development as AI has become more of an offensive tool within cyberattacks. “AI is changing how cyber operations are built and delivered, increasing the speed and sophistication of attacks,” he says.

In particular, “threat actors can automate the discovery, weaponization, and delivery process at a scale we haven’t seen before. Accessible GenAI tools are also lowering the barrier to entry for attacks, making it easier to generate targeted phishing lures and convincing synthetic media for impersonation and social engineering.

“AI is making established threat actor techniques faster, more adaptive, and more scalable,” Darrah says.

How Businesses Can Navigate 2026, and Beyond

Navigating today’s landscape can be disorienting, and Darrah acknowledges this. “We are living in strange times where reports of nation-state meddling, ever changing privacy laws, advances in technology such as AI, data extortions, and data breaches are constant.”

His advice for small businesses is to prioritize strategically. Not everything is equally important, and he recalls the earlier advice of a colleague when he says, “if everything is a priority, nothing is a priority.” Somewhat cliche, he adds, but it resonated with him.

In practice, Darrah’s advice is to follow the basics, particularly if you’re tight for security budget. “We live in a world that values new and novel, but cyber criminals exploit the basics of network security hygiene and social engineering awareness.”

Faylo, on the other hand, recommends similar. “For budget-constrained clients, we generally advise them to fix the ‘insurability gaps’ first.’ These are multi-factor authentication, endpoint detection and response, and ensuring you have restorable and tested system backups.

“The common thread is that these are governance and process failures, not just technology gaps, which is why we push clients toward treating cyber resilience as a leadership level priority, not something delegated entirely to IT.”

It’s no surprise that the more of your company exposed to best cybersecurity practice, the stronger your company will be.

This article is part of interviews with Linda Yao, Vice President and General Manager of Hybrid Cloud & AI Solutions at Lenovo, and Michael Benisch, Vice President of AI at Motive. These interviews were conducted for Tech.co’s AI newsletter, The AI Strat. For more interviews with AI experts on the latest trending topics, subscribe to the newsletter.

Undoubtedly, AI has developed fast. But, so far, progress has been mostly confined to a laptop or smartphone screen. Chatbots and agents are more powerful than before, workflows can be fused with AI for stronger efficiency, and so on.

On the physical side, progress has been slower, which isn’t unusual for an industry that usually takes a longer time to integrate new technologies.

Despite this slower pace, physical AI is still changing the way businesses are operating. AI within warehouses and transportation has developed substantially, increasing productivity across the board. But experts say physical AI has benefits beyond productivity, potentially making whole industries and professions safer for human operators.

I spoke to Linda Yao, Vice President and General Manager for Hybrid Cloud and AI Solutions at Lenovo, and Michael Benisch, Vice President of AI at Motive, to see where industries are at now with physical AI and what we can expect for the future.

Physical AI Has Been Slower to Develop

In a 2025 keynote, NVIDIA CEO Jensen Huang claimed: “The next frontier of AI is physical AI.” When I first start talking with Michael Benisch, VP of AI at fleet management solutions provider Motive, he uses similar language to talk about physical AI, calling it “one of the last frontiers,” that needs to be unlocked in the AI world.

Benisch has been working on physical AI for almost a decade. His resume includes the self-driving division at Lyft, as well as stints at Toyota and General Motors.

“When people think about AI, it’s usually the chatbots we talk to on the web or if you use Claude Code or something like that,” he says. “But then there’s the other half of the world out there, where you need to get your packages from Amazon, and you need to get your real food.”

When I asked him why progress was generally slower, he said there are a lot more obstacles you can’t overcome, when you’re dealing with the physical world. Unlike computer programs, issues can’t be fixed by “simply hacking through it in a weekend.”

How Organizations Have Been Adopting Physical AI

Plenty of companies are already seeing massive gains with physical AI. Linda Yao, VP and General Manager of Hybrid Cloud & AI Solutions at Lenovo, helps organizations adapt to emerging technologies. At one point, this was hybrid cloud, and now it’s AI.

One company, ST Logistics, wanted to, in Yao’s words, “automate as much of their warehouse as possible.” The formula for this case was physical plus digital AI, Yao explains. On the physical side, this includes robots, and they sound like the kind of robots you’d expect in a typical sci-fi movie.

One robot helps “efficiently move the equipment or move the inventory from one side of the dock into the receiving inventory warehouse,” and another does “the physical loading and unloading.” Then, there is another robot that is more “detailed,” it identifies the right boxes and stacks them in the right order.

For the digital side, Yao explains: “We are working to make sure that the systems being used, the IT systems being used to log the inventory, to track the data, to take the orders, to make sure that the orders are shipped on time. That’s all then connected back into not just the human workflow, but also the robot workflow and the AI workflow.”

Overall, Yao says ST Logistics reduced order processing time by 40%, reduced energy consumption by 30%, and increased productivity by 30%. “It really is bringing together the physical as well as the digital, as well as the human element to make this AI solution work.”

Physical AI Can Make Physical Professions Safer

Yao describes the kind of work going on in warehouses as “backbreaking,” particularly for workers that are loading and unloading, or spending an hour in cargo containers that can get to 120 degrees Fahrenheit in the summer. A lot of the time, she says, companies are struggling to find people who even want to do that work.

This obviously makes the logistics industry a prime candidate for AI. Especially since trucking is considered one of the most dangerous careers out there: according to the US Bureau of Labor Statistics, almost 800 truck drivers lost their lives in a work-related incident in 2024.

Motive’s suite of AI-powered products offer increased safety for drivers out on the road, Benisch tells me. One company, Agmark, adopted Motive’s AI dash cams, and as a result, saw a 67% reduction in accidents and a 15% reduction in unsafe driving behaviors. 

Similarly, chilled products distributor Damian’s Enterprises saw a 20% improvement in safety scores when they adopted Motive’s AI Dual-facing dash cams.

Improving safety, Benisch explains, is one of the ways fleet operators and managers can get their drivers behind them when AI is implemented. He encourages leaders to tell their team, “You have the ability now to do a better job than you did before. You can be safer. You’re more likely to come home at the end of the shift.”

What’s Next in the Physical AI Space?

At the end of our interview, I asked Benisch for his physical AI predictions in the next year. A question I sometimes feel cruel for asking, especially since one year’s progress can take place in a few days with AI.

“Physical AI isn’t moving as fast as the other AI areas… it’s not like the rocket ship growth that you see from ChatGPT or Claude Code,” he says. He does predict, however, “there will be market differences and things that are hugely improved.” Within the driving space, he predicts “the passenger space will be more and more automated, but taxis will still be there, and Lyft and Uber will still have human drivers.”

Like Benisch, other experts say the same about human contribution within the physical AI space.

“Between now and 2040, I certainly expect robotics and physical AI to create at least a trillion dollars in economic value, most of that in manufacturing and logistics. That will happen not through worker replacement and productivity but through our ability to innovate what products we make, how we make them, and what tasks humans perform versus which ones are automated, elevating the nature of work for humans to more problem-solving and to sort of a catcher role, rather than just relying on our physical dexterity.” — Ani Kelkar, in an article for McKinsey & Company

While we’ll certainly see the most gains from physical AI within logistics and manufacturing, this could lead to physical AI developments across other industries. Bloomberg reported earlier this month that OpenAI are planning to release their first hardware device, a screenless speaker with integrated AI capabilities.

The design of the product is still currently under development, but is being pitched internally as a “humanlike AI companion that lives in the home.” Though fascinating, I think most would prefer a robot that carried groceries from the store.

Adam Darrah, Vice President of Intelligence at ZeroFox, says most ransomware attacks are about building notoriety and embarassing organizations, rather than just for financial gain.

North American-based organizations were the most targeted by malicious actors in ZeroFox’s latest ransomware report, and industries including manufacturing and construction were hit the hardest.

Darrah says organizations should consolidate cybersecurity fundamentals to protect themselves, although admitted the firepower of nation-backed groups could be too difficult to stop in the long run.

Most Ransomware Attacks are About Embarassing Organizations Over Taking Their Money

ZeroFox’s Q2 2026 Ransomware Wrap-Up found five attack collectives were responsible for nearly half (49.5%) of all global ransomware and digital extortion (R&DE) attacks in Q2 of 2026.

Adam Darrah, Vice President of ZeroFox, spoke to Tech.co exclusively about how R&DE attacks are primarily about embarassing and panicking organizations, rather than financial gain, based on the findings of the report.

“What [the findings] suggests is that it’s less about financial incentive and more about appearing bigger, better, scarier. They want the civilized, law-abiding class to be scared and to be embarassed,” Darrah told us.

Manufacturing, Supply Chain, Technology Among Industries at Risk

The report identified a total of 1,885 separate R&DE incidents in Q2 of 2026, an increase in year-on-year incidents from Q2 2025 and Q2 2024, suggesting incidents around this quarter are on the rise. Notably, attacks on North American-based organizations made up nearly 45% of all incidents reported. According to Darrah, these numbers are the result of the geopolitical conflict between America and Iran.

Overall, the manufacturing industry experienced the highest number of R&DE incidents in this quarter, accounting for nearly 20% of all incidents. This too, Darrah says, is connected to the shutdown of the Strait of Hormuz, and a desire from ransomware groups gain notoriety.

“Ransomware groups, especially those aligned either as a fellow traveler or as a direct associate of a military security apparatus, they like doing a couple of things. They love notoriety, they love being obedient to their overlords, they love money,” Darrah explains. “But they also love sitting back and watching civilized, law-abiding people and organizations squirm and be uncomfortable and to watch that pain.”

Similarly, R&DE targeting patterns have shifted away from the retail sector, and have instead turned towards the technology industry. Darrah again notes this is another way of embarassing important organizations.

“There’s a lot of prestige with what’s happening in the world right now in terms of AI,” Darrah says. “I think there’s probably a bit of prestige in going after technology firms, the people associated with that, to embarass the company and the victims.”

While Darrah notes developments in AI have also made the threat landscape more unpredictable, he does say some ransomware groups are still using older technologies, and are yet to convert to AI. In this case, businesses should protect themselves from all kinds of threats.

What Should Businesses Be Worried About?

“I think businesses should be most worried about worrying too much,” Darrah says, suggesting businesses may be overcomplicating the issue. “I think we overthink the latest and greatest, and we take our eye off the fundamentals. And if you concentrate on the fundamentals, that is going to significantly decrease your risk.”

Darrah urges businesses to focus on strong password policies, a culture of reporting suspicious emails, and patching, among other best cybersecurity practices.

However, there are some vulnerabilities businesses can’t avoid. Businesses today are frequently outsourcing to clouds and third-party providers, and yes, they make our lives easier, but if they are breached, it becomes a problem. Darrah says this “affects people who are doing everything they’re supposed to do. They have all their policies… But if that third party gets hit, it’s like your brand is still dragged through the mud.”

“The interconnectivity of the world just makes it ultra likely that somebody is going to be touched by this plague,” he adds.

Moreover, sometimes organizations are targeted by much larger and more sophisticated groups than they can realistically handle. Particularly if they are backed by nations like Russia or China.

“I think we put a lot of pressure on people who love their job and who care very deeply, and we’re telling this person, defend against Russia, defend us against China,” Darrah says. In these cases, the outcomes can be much harder to prevent against, and understandably so.

Workers were more likely to turn to an AI agent for the answer to an obvious question than their own manager, in a new study.

While AI is changing hybrid, on-site, and remote workers differently, the study finds current company-wide training schemes may be insufficient, and workers are turning to other education tools.

Managerial roles and general workplace structures could change as workers turn more and more to AI, but the study shows we may not be at a place to trust AI completely yet.

Workers Would Rather Ask AI than Managers

In a new study from Adobe Business, 68% of workers reported turning to agentic AI for answers to easy questions, compared to only 4% that ask their manager. Overall, the study found workers were 17 times more likely to ask an AI agent questions they’d hesitate to ask leadership.

Similarly, more than half (54%) of respondents prefer using agentic AI to draft difficult or high-stakes messages, compared to the 23% that turn to managers or peers. Plus, 62% turn to AI for recapping an important meeting, versus 10% that turn to managers.

These findings suggest AI is becoming a trusted, judgement-free colleague for many workers. The study surveyed over 1,000 full-time employees who use AI weekly, with the goal of understanding how agentic AI has become part of every day tasks.

Agentic AI Training Is Lacking

According to the study, the next wave of AI, agents, are here, with 92% of AI users having adopted agentic AI at work. Workers, however, are using AI differently based on their work location. Hybrid and onsite workers mostly use the technology for idea generation and brainstorming, whereas remote workers use it for content creation and drafting.

Similarly, nearly three in four (73%) of workers say AI has improved their job performance, suggesting workers’ confidence as well as their efficiency has been positively impacted. 50% of workers also report freeing up time for higher-value work, and 38% say it has improved productivity.

However, the study finds a steady gap across all job levels when it comes to AI training. When asked the ways they have learned to use agentic AI, over half (54%) of workers said they had used trial and error, and 36% said YouTube. Meanwhile, only a quarter of respondents said they’d learned through collaboration.

That isn’t to say companies aren’t providing any training at all, it just may be the case that training programs aren’t providing all workers need or want to know. In fact, the study found one in three users had received formal agentic AI training, but fewer than half of those trained have gone on to build their own custom agent.

How AI Could Shift Workplace Structures

There’s a convincing argument to be made around AI shifting corporate structures. With access to more advanced tools, workers may be able to complete some tasks more quickly than their managers. Likewise, as AI skills become more and more desirable within roles, those with more knowledge may find it easier to make an impact, even if they have less experience.

Where we might also see significant change is for remote workers. Compared to on-site workers in the study, remote workers saw 17% higher productivity and a 35% greater reduction in imposter syndrome.

Traditionally, remote workers don’t always have access to managers or employees to check in with. AI could help provide this support, potentially improving employee wellbeing and lowering work stress.

However, despite these positive findings, it’s unlikely we’ll see AI replacing any structures yet. The study found 43% of workers prefer to check a risky idea with a colleague, compared to 39% who would turn to AI. This suggests trust and reassurance is still best received by a human colleague, and that AI may be too young in its current iteration for workers to fully trust its judgement.

AI is no longer just making cyber attacks cheaper and faster, but it’s now in charge of running whole operations, a new study has found.

Cybersecurity leaders have been warning about the dangers of AI for some time, and developments such as a maturing AI-enabled criminal tooling market and more sophisticated imitations of voice, face, live video, and documents pose serious concerns.

Experts are urging businesses to integrate cybersecurity practices within organizations, including cybersecurity leads and teams. Likewise, businesses should align their security protocols with the AI applications they are using to have a foundation to build from as threats continue to change.

Study Finds AI is Now Operating Attacks on Behalf of Users

In Check Point’s 2026 AI Security Report, AI has taken a leading role in cyber attacks. The technology is no longer just enhancing cyber attacks by making them cheaper, faster, and more accessible. Instead, it’s now doing the hands-on operational work itself.

In one example outlined, a human attacker carried out a breach by running two commercial tools, Claude Code and GPT-4.1, together. This led to the breach of 9 Mexican government agencies.

The report tracks how AI has reshaped the threat landscape over the past year, and is grounded in real-world incident and Check Point Research threat intelligence.

How AI Threats Have Evolved in the Past Year

Check Point found AI is now able to turn a new system vulnerability into a working exploit within hours, giving companies little time to react.

“[We] hear from the security operations center and incident respond teams [that] if a human was doing these tasks, we’d at least have some time to detect it, respond, and try to prevent it, but the agents are so fast, by the time [we] know about it the damage is already done.” Adam Ely, general manager of AI Security at Check Point Software, during a media roundtable at Check Point Engage Singapore 2026

Similarly, existing AI threat tools have matured in the past year. Phishing-as-service kits now have language models with jailbreak features built-in, and conversational AI voice-agents can run vishing and one-time passcode attacks.

Furthermore, the report notes that Virtual Identity measures can no longer be fully trusted. Thanks to AI, attackers can imitate voice, face, documents, and live video with ease, which makes communication scams more widely available across multiple platforms.

Protecting Your Business in 2026 and Beyond

Businesses are increasingly at risk as AI becomes more common in cyber attacks. The study found 1 in 17 AI prompts carry a serious data exposure risk, a particularly concerning statistic for organizations typically running 10 AI apps a month on average. Many of these AI apps, the study notes, also remain ungoverned.

Overall, high-risk prompts doubled to 4% in a year, one example of how AI is changing the cybersecurity landscape in a rapid fashion. “Enterprise data leakage through GenAI is a persistent and growing risk,” the report warns.

Not only are organizations going to have to ensure their core security principles are intact and solid, but they’ll have to continue to rework and update these principles, in order to keep up with new risks.

“Ensuring that companies are putting their protections for AI really in line with the AI applications or at the runtime level of the application is really important. As models change or functions change, that implementation can then be built on versus having to do a completely new security implementation.” Adam Ely

Adam Ely, general manager of AI Security at Check Point Software, believes organizations should also ensure cybersecurity teams and leaders are fully integrated into business operations. With these teams, businesses should have regular conversations about potential threats and how these are changing.

82% of business leaders say they are experiencing barriers preventing them from making the most out of AI, according to a new study.

The report finds that AI adoption has, however, been a positive investment for small businesses. Over half of owners said they had seen return on investment from AI, and almost half of owners saving at least 4 hours per week.

On the other hand, security and a lack of trust are the biggest barriers preventing owners from fully exploring AI. Many leaders aren’t yet confident letting AI handle low-level business tasks without human supervision.

Over 80% of SMB Owners are Experiencing Barriers with AI

In a new study from financial technology company Bluevine, leaders reported overall optimism on AI. 68% of SMB owners said they believe AI advances will help their business, up from 61% from the year above.

While sentiment is high, however, 82% of owners are encountering at least one barrier that is preventing them from exploring AI more deeply, and potentially using the technology for all it can offer. This suggests leaders aren’t yet reaping the full benefits, even after investing in and deploying the technology.

AI is moving fast, and the study suggests it may be moving faster than leaders can. Findings come from 942 small business owners and financial decision-makers across the US.

Over Half of SMBs are Seeing Return on Investments

Over half (52%) of business leaders reported seeing return on investment from AI, compared to only 24% that said they haven’t seen any gains yet. This could mean many businesses are still in the trial and error phase, testing out AI tools to see which ones benefit them the most.

On the other hand, other studies have suggested communication on whether AI has been beneficial can be fragmented across companies, leading to some uncertainty on whether AI has benefited a business or not.

Beyond ROI, business leaders are seeing gains across the board. Almost half (48%) of SMBs report getting back at least half a workday every week when using AI. 23% reported saving four to six hours per week, and 11% reported savings of seven to nine hours per week.

Distrust and Data Security Remain Big Concerns

Adoption is high – the study saw it rise 50% in just six months between September 2025 and March 2026. But, businesses are experiencing barriers beyond initial deployment. A significant 82% of leaders report at least one barrier to using AI deeply, arguably preventing them from taking full advantage of how the technology can help business operations.

33% of leaders cited data security concerns as their biggest barrier, up from 23% the year before. There remains among leaders a lack of understanding as to what these new tools are doing with their data, suggesting more training is needed to avoid costly data breaches.

Similarly, 31% of owners said a lack of trust in AI’s accuracy was another barrier. Only 22% of owners across the study said they were completely confident AI could do low-level business tasks without human supervision, with 78% not fully trusting AI in this sense, even when these tasks were those that would save them the most time.

Productivity, time savings, and ROI remain important metrics for businesses to focus on when using AI. However, this trust factor seems to be preventing businesses from reaping the most benefits.