Gerard Grouve (@Grouve) header
Gerard Grouve (@Grouve) profile picture

IT Manager & Strategic Advisor | Orchestrating Sovereign Intelligence for AUM Entities | Specialist in System Innovation & Operational Excellence | Le Mans Winn

Nederland
Joined May 2008
Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)Photo by Gerard Grouve (@Grouve)
Let agents earn their autonomy. AT&T runs almost 500 agentic workflows. Each starts under human oversight and only goes autonomous at mid-to-high 90% accuracy. Great model. One addition: that score belongs to one model, so re-measure when you switch. siliconangle.com/2026/11/14/…
6
Your suppliers are part of your attack surface. At least 14 Dutch websites served malicious ad code through one compromised ad platform, aimed at iPhones on iOS 18.7.2 or older. Easy win for IT teams: enforce a minimum iOS version on business phones. nos.nl/artikel/2639562-meerd…
12
Model retirements deserve a calendar slot. Claude Sonnet 4.5 retires on 30 November. Since API version 2026-10-21, temperature, top_p and top_k return a 400 error. Good habit: check which keys still call it and re-test prompts on the successor. platform.claude.com/docs/en/…
12
Gerard Grouve retweeted
Mais imagens do acidente.
237
1,629
208
8,987
977,941
Stel: je moeder krijgt een spraakbericht dat precies klinkt als jij. “Ik zit in de problemen, maak snel geld over.” Spreek vandaag een familiecodewoord af. Geen codewoord? Ophangen en zelf terugbellen. Lees Echt of nep? (€9,99): payhip.com/b/08cmR
1
18
Stel: de cloud ligt eruit. Het noodplan staat in diezelfde cloud. Test vandaag: wie belt wie, wat moet doorgaan en waar staat de offline noodkaart? Een plan is pas bruikbaar als je erbij kunt. Lees Als alles uitvalt (€29): payhip.com/b/jvZNy
7
Give every AI key a spending limit. Railway estimates stolen customer AI keys ran up $740,000 in AI usage. The attacker hunted for AI provider keys first. Quick fix: a hard monthly cap and usage alert on every key, and know where each one lives. thehackernews.com/2026/11/ra…
8
The stop button belongs outside the agent. OpenAI checks every agent request with a central service. A stopped job gets no output within 35 seconds. If that service fails, everything closes. Great idea to copy: test how fast you can stop yours. openai.com/index/control-roo…
9
The future isn’t a wave to survive—it’s a course to steer. Stop waiting for the ultimate tech tool. Grab a single sheet of paper and answer three questions: Where do we lead? Where do we draw the line? What must our people learn? Clarity brings control. 🚀 #AI #Leadership #SME
1
17
Old TODOs now cost real money. Google found that coding-agent sessions chasing stale TODOs used 77% more tokens and 105% more wait time. 41% chased bugs already closed. Great news, it is cheap to fix: clean stale TODOs before your agents read them. teapawt.org/glenngabe/status/21190…
From a Google software engineer and data scientist. Your TODO comments could be distracting agents -> Agent Distraction "As agents' agency through tools and skills grows the ability for side-quests scales proportionally. Obsolete specs, long resolved bugs, and lagging documentation no longer rest in long forgotten corners of the codebase, but rather are unintentionally ingested as context indirecting and misdirecting the next wave of coders - the agents. This indirection is not free: in the best case it only costs tokens, steps, and time - in the worst case it distracts and grounds the agent on falsehoods leading to incorrect logic and decisions during implementation." And what they did about it: "As a result of this multi-layer approach, our todo cleaner has been able to remove thousands of actually obsolete references over the last 6 months exceeding 95% acceptance from human reviewers." developers.googleblog.com/ag…
2
16
Gerard Grouve retweeted
One of the greatest and most American pace laps ever!
47
461
39
3,486
155,908
Gerard Grouve retweeted
Those who have guardian angels
117
1,720
71
18,182
1,520,171
HUGE crash for the #35 Alpine! 💥 Charles Milesi goes off and makes contact with the barriers. He has safely exited the car, but the #35 Alpine has been forced to retire from the race. #WEC #6HMonza #Alpine
8
91
30
633
46,823
Lausitz Rallye standings after ss 2 #drm #lausitz #rallye #rally
1
29
Without shared meaning, AI keeps guessing. Franklin Templeton only got useful research agents after they built a semantic layer — what the data means and how it connects — before scaling. Define meaning first, then let the agents run. openai.com/index/franklin-te…
13
Your cold chain is an attack surface. FBI and CISA warn of attacks on unpatched, internet-facing refrigeration controllers. European waves hit ~250-350 French stores, then over 115 stores in six countries. Pull controllers offline. Change defaults. Treat OT as production.
18
OT blind spots are a board issue. Palo Alto Networks surveyed over 1,600 critical infrastructure leaders: 68% lack complete real-time OT asset visibility, and 95% worry about frontier AI-powered attacks. Map every connected OT asset first. Then automate where downtime hurts.
15
Patches at machine speed. In one run, an OpenAI pipeline validated and patched 948 Linux kernel privilege escalation flaws, at about $1,279 each. Defenders now have that speed too. Make sure your patch process can keep up. teapawt.org/OpenAI/status/21181194…
Our security researchers built an autonomous cyber vulnerability discovery, exploitation, and patching pipeline. In a single run with an upcoming cyber model, it validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel at a total API cost of $1.2 million.
23