Microsoft Patch Tuesday in July set a record by addressing 722 common vulnerabilities and exposures, including active exploits in Active Directory Federation Services and SharePoint. The update also marks the end of support for SharePoint Server 2016/2019 and SQL Server 2016. The update, KB5101650, brings version 25H2 to build 26200.8875 and version 24H2 to build 26100.8875.

AI, particularly Microsoft's Multi-Model Agentic Security Harness, helps identify vulnerabilities faster, resulting in larger security updates. Microsoft emphasizes that this does not indicate a decline in security but rather an improvement in vulnerability management.

The AI boom is reshaping the CPU market, attracting new players such as Nvidia, Qualcomm, MediaTek, Arm, Google and Amazon, as well as Chinese companies. The market for CPUs in AI data centers is expected to grow more than 40% annually, surpassing $220 billion by 2030. This surge in demand is driven by the increasing need for CPUs to handle AI workloads, particularly agentic AI, which requires more orchestration tasks. Traditional leaders such as Intel and AMD face competition from companies that leverage Arm-based architectures for energy efficiency.

Recent research shows that generic AI productivity tools can inadvertently increase workload, produce workslop or pose security risks, thus wasting time, eliminating return on investment and diminishing critical thinking. Leaders can avoid these problems without banning AI if they adopt clear usage policies, provide sanctioned alternative tools, train employees to verify output and ensure email authentication, among other steps.

OpenAI's GPT-5.6 offers incremental improvements over GPT-5.5, particularly in code review and implementation tasks. The model comes in three sizes -- Sol, Terra and Luna -- and features varying reasoning levels, with higher levels consuming more usage limits and responding slower. Users are advised to use extra high reasoning for planning and medium for implementation to optimize performance.

Businesses are turning to lower-cost Chinese AI models for tasks such as coding, customer support and internal automation, driven by cost savings compared with US-developed models. Companies including Airbnb and DoorDash have adopted Chinese AI models. "Many companies do not need the 'Ferrari' of AI for every task; instead, they are choosing lower-cost models that can handle large, routine workloads at scale," says Eugene Cheah, CEO of AI platform Featherless.

Despite stricter return-to-office mandates from companies such as Amazon and AT&T, remote work remains prevalent, with 35% of full-time employees working from home at least part-time in 2025, according to the Bureau of Labor Statistics. This trend persists as office attendance reached post-pandemic highs, and experts continue to advocate for hybrid work options amid rising employee burnout and workplace changes driven by AI.

The US labor force is projected to shrink by nearly 6 million workers by 2032 due to declining birth rates and Baby Boomer retirements, according to Indeed Hiring Lab. The demographic shift will be troublesome for sectors such as healthcare, construction and skilled trades -- while AI is expected to automate many tasks, it cannot replace the human labor needed in these fields. Employers will need to invest in apprenticeships and training to fill critical roles.