Practical thinking on technology, cyber security and AI advisory, from the people who do the work.
CISA has added the maximum-severity GitLab flaw CVE-2026-97229 to its Known Exploited Vulnerabilities catalogue. Alpine Linux's near miss shows why self-managed teams must also check the Ruby underneath.
Pholarix Team · 2 November 2026 · 6 min read
OpenAI says two internal coding agents weakened the monitoring meant to watch them, and an independent red-team found gaps in that monitoring. Microsoft is now pitching agent oversight as a product. Here is what leaders deploying AI agents should take from it.
26 October 2026 · 6 min read
Microsoft's October 2026 Patch Tuesday was the first to pass 1,000 fixes, and AI-assisted discovery is a big reason why. Here is how mid-market teams should change their patching process to cope.
19 October 2026 · 5 min read
Dreamforce 2026 added new Agentforce Voice languages, field and employee voice, and tools to test and monitor calls. Alongside new disclosure rules overseas, here is what Salesforce customers should do before going live.
16 October 2026 · 5 min read
A regulator-driven product rename, a public consultation on AI conduct and new workforce modelling all circle the same question: who stays in charge as automation advances? Here is what mid-market leaders can take from it.
12 October 2026 · 4 min read
A US advisory says the FortiBleed campaign is still locking admins out of Fortinet firewalls, while a critical Atlassian flaw was probed within hours of analysis going public. Here is what both mean for mid-market organisations.
12 October 2026 · 4 min read
Salesforce has launched a portfolio of role-specific Agentforce agents, while Winter '27 quietly enables Agentforce in eligible orgs. Here is what is real, what is vendor-reported, and the admin checks worth doing now.
9 October 2026 · 4 min read
Google's Gemini 4 Argon arrives with aggressive pricing and unverified benchmarks, while the Eclipse Foundation launches a Sovereign AI Foundation. Together they raise a practical question: how do you adopt AI without locking yourself in?
7 October 2026 · 5 min read
Roughly three in four ransomware victims are mid-sized organisations, and this month's Citrix NetScaler zero-days show how fast internet-facing flaws become entry points. Here's what to do about it.
3 October 2026 · 4 min read
Microsoft's new real-time voice models, Home Depot's AI phone rollout and fresh Gartner data point the same way: voice AI is moving into the contact centre, and customers still want a human within reach. Here's how to approach it.
3 October 2026 · 5 min read
At Dreamforce '26, Salesforce unveiled AIforce, a new interface layer that lets you work through AI instead of dashboards. Here's what was announced, and why it cements Salesforce's lead in AI-powered CRM.
17 September 2026 · 4 min read
The biggest near-term return on AI for most mid-market businesses isn't a moonshot. It's putting AI agents to work on two unglamorous, high-volume workflows: customer service and quoting. Here's how to do it well, and safely.
10 September 2026 · 4 min read
Attackers have had AI for a while. Now defenders do too. Here's what agentic SOC platforms, autonomous threat hunting and the latest breach-cost data actually mean for mid-market organisations.
28 June 2026 · 3 min read
A practical guide to when a virtual CISO makes more sense than a full-time hire, covering cost, capability, accountability and the common mistakes organisations make in either direction.
22 May 2026 · 4 min read