An unnamed enterprise spent $500M on @ClaudeAI in a month after failing to put usage caps on employee licenses.
Most are reading it as "cap AI spend." The actual lesson is upstream: spend is not the metric. Engagement is.
🎙️ @TheGenAICEO on @section__ai
https://t.co/AqqF7jTxsv
Some conference sessions end. Others keep moving after the room clears.
Last week at @FortuneMagazine#BrainstormTech 2026 in Aspen, @TheGenAICEO joined the Fortune Tech Breakfast: How to Rebuild Society for an Intelligent Age.
This was the second kind. 🧵
This exact agentic shift is hitting IT infrastructure right now. NetOps teams cannot afford to spend hours manually triaging traffic bottlenecks or digging for missing security patches.
It's why modern enterprise tools are baking autonomous intelligence directly into the dashboard. Solutions like @GFISoftware’s #Exinda Network Orchestrator and #LanGuard now feature AI-powered features. They act as autonomous assistants within your console, handle predictive troubleshooting, auto-suggest bandwidth-optimization policies, and deliver tailored patch-configuration advice in natural language.
The future of work is delegation!
The way we handle knowledge work just shifted permanently. Brand new research from @perplexity_ai and @HarvardHBS shows a massive 48x explosion in machine execution time when users switch from traditional search to autonomous #AI agents.
We are officially transitioning from manual data collectors to high-level system supervisors. Instead of clicking through a dozen tabs to stitch together information, the agent handles the multi-step grinding over hours-long horizons while humans verify the final output...
The operational efficiency numbers from this field study are wild. By letting an agent orchestrate the task decomposition, the Human + AI workflow chopped average task completion times down from 269 minutes to just 36 minutes. That is a massive 87% reduction in time and a 94% drop in estimated labor costs.
Research, deep data analysis, and asset creation took the largest slice of agent usage. Dive into the complete Harvard/Perplexity research document here to see the full data:
https://t.co/Uy47e02OWO
DLP rules by hand: 40 hours.
DLP rules from a plain-language policy description: 5 minutes.
GFI MailEssentials AI. Available in version 22.4+.
See what is included: https://t.co/jTn1L4lTx6
#GFIMailEssentialsAI#DLP
ChatGPT is getting better at remembering what matters: your preferences, constraints, and the context that helps you pick things up where you left off. And with memory summaries, you can review and steer what it remembers.
Rolling to all users over the next few weeks, starting today with Plus and Pro users in the US.
This is exactly why setups like Kerio Control and MailEssentials from @GFISoftware are so critical!
With Kerio's AI-driven Shield Matrix and MailEssentials AI' advanced threat scanning, you get the automated, behavior-based defense needed to stop AI attacks at the edge.
Simple filters and static blocklists will not stop an AI-driven threat that adapts on the fly. Security teams need to pivot to behavior-based tracking that catches weird activity the exact second a threat breaches the perimeter.
What is your team doing to adapt to this?
#NetworkSecurity #SecOps #KerioControl #ShieldMatrix #MailEssentialsAI
Cyberattacks just got a massive #AI upgrade, and standard security setups are officially falling behind.
A new @AnthropicAI study shows how hackers are using AI to automate complex, multi-stage attacks deep inside networks.
Instead of just sending basic phishing emails, these new threats string together autonomous moves to bypass classic firewalls.
If your perimeter defense relies on old, static rules, it is time for a serious reality check.
#CyberSecurity #Automation
The scariest part of this AI threat shift is how attackers handle lateral movement and network discovery.
Legacy security tools just are not built to spot these smart, connected actions. We need intelligent updates that can flag these automated chains before they spread.
Check out the full breakdown from the report here to see what we are up against: https://t.co/njD1cExLxt
#InfoSec #TechTrends