Teams | Collaboration | Customer Service | Project Management

OpenAI's new Apple Messages plugin gives ChatGPT access to the Messages application on a Mac. Once enabled, users can ask ChatGPT to find conversations, summarize them, draft replies, and send messages. That has understandably raised questions about what it means for iMessage privacy. The starting point is understanding what the plugin does and, just as importantly, what it doesn't do.

Employee productivity is calculated by dividing total output by total input. When time is the input, use output divided by total hours worked.

Every project plan opens with a number nobody checks again. A task is quoted at six hours, work begins, and by the time it wraps at eleven hours, the gap has already been absorbed into the next deadline instead of the next estimate. A task time tracker exists to close that loop. It turns every completed task into a reference point, so the next estimate is built from what actually happened rather than what felt reasonable at the time.

Email is one of the few workplace tools that creates more work simply by being available. Every message asks for a small decision: open it now or later, reply or ignore, archive or keep, delegate or handle personally. None of these decisions seems particularly demanding on its own, but together they can consume a surprising amount of time and attention. The problem is not always the number of messages.

IT support has a familiar problem. An employee says, “I can’t access Salesforce,” but solving that request may involve much more than answering a question. Someone has to understand what access is needed, check whether the employee is eligible, collect missing information, trigger an approval, update an identity platform, record the request in the ITSM system, and tell the employee when it is complete. For years, service desk chatbots solved only the first part of that journey.

AI makes it quick and easy to turn rough notes into a clean, easy-to-read document. Job well done, right? Think again. Atlassian’s Teamwork Lab suspected that this professional veneer—what we call “AI polish”—makes it harder to spot foundational flaws and give feedback, so we tested it. We wanted to know: Does AI polish make it harder to identify problems in early drafts? How can we solve for this blind spot?

As an owner or managing director, you often assume ops responsibilities by default. But that’s a short-term solution. One that inevitably leads to problems and burnout as the business grows. And being overwhelmed often leads to rushing your first ops hire. Then, instead of being able to fix problems, your new team member gets stuck firefighting. Hire strategically instead, with our seven-step process.

The office perimeter is gone. Work now happens on laptops in living rooms, phones on home Wi-Fi, USB drives passed between desks, and screens left unlocked in co-working spaces. None of that sits behind the firewall your IT team spent years building. Most security budgets still protect the network far better than they protect the device at the edge of it, and that gap is where endpoint security risks quietly build up.

AI is absorbing the parts of your team’s work that used to prove someone was doing their job. If you manage a distributed team, that leaves a real question: what do you actually measure now, and how do you know it’s true? The location debate is already settled. US remote work has held steady at around 27% of the workforce for three years running, according to figures Liam Martin, co-founder of Time Doctor, cited in his opening keynote at Running Remote 2026.

ServiceNow AI is no longer just about Virtual Agent. Today, its AI stack includes Virtual Agent, Now Assist, AI Agents, AI Agent Studio, and Otto—covering everything from conversational self-service to autonomous task execution. For IT and HR leaders, the real question is not whether ServiceNow can use AI.