Last night, I watched for the first time, a video where kidnapped school children were being tortured by bandits with tongues of flame from a burning plastic. I’ve never been deeply disturbed in a long while like I was last night. For the first time I wept for Nigeria. This is totally unacceptable. Why on earth will a nation allow such level of violence against children? To what end? Where is the state of emergency? Where is the surveillance helicopters combing the forests? Where’s Tinubu’s comment on this incident? Where is Nigeria 🇳🇬?
Apart from renting, you can easily afford to buy quality houses up north with, sizable rooms, driveway and large garden. A colleague got a 4 bedroom, two bathrooms, with driveway and large garden, semi detached in L5 area of Liverpool for 190k last year and the house is currently valued at £220k. That type of house should be around £500k and above in most southern parts of England
@ochapasamuel1@mrboboskie Bournemouth is not Northern England. Besides it’s expensive there. If you want to try up north, consider Liverpool, Wigan, Leeds, Bradford, Newcastle, etc
It’s not about just shouting @PeterObi and jumping around the man in public places. Do you have your PVC? Have you set in motion, a grass root mobilisation plan to help people get their PVC?
After the tragic incident that befell GUO last week, it's time transport companies hired private security outfits to travel with them on the road. Sounds like a long call but given the security issues in Nigeria and the Government's lack of commitment to tackling them, we must think of better ways to keep road travellers safe, regardless of how unconventional that may sound. @ARISEtv@ruffydfire@guotransport
@Evanschizzy9708@OurFavOnlineDoc When you say "electing", it's as if we actually voted her in. No. That's not the case. They rig their way in. So most of them in the house weren't elected but selected by their ogas at the top. The sooner we get INEC sorted, the better for us all.
@Amaka_power@prophetswitch He wasn’t voted in the first time. You guys should be mindful of that language. He rigged his way in and he’s got every apparatus in place to repeat that, sadly.
Apart from the long sponsor reads (which take ages), Lex Fridman's podcast is one of my favourites. Anyways, his ads style, his choice.
This weekend I listened to his chat with Peter Steinberger, the developer behind OpenClaw.
Days after telling us his plan was "security first, then ship to everyone," Steinberger joined OpenAI.
I don't subscribe to the notion that he shouldn't have joined Altman.
What concerns me really is what happens to the security-first promise now that he's handed OpenClaw to a foundation while building agents at OpenAI.
We've seen this before with the very company he joined.
Altman's founding vision was to make AI accessible by keeping it open source.
Today, OpenAI has restructured away from its non-profit roots, with less openness than originally promised.
My focus in this post is not to dole out a stream of invectives on individual choices. Far from that.
My submission is that organizations cannot take vendor promises at face value when deploying AI systems touching their data, customers, and regulatory obligations.
They must do the dirty work of due diligence and evaluate which AI agents offer not just ROI but the lowest risk threshold for their specific context.
Because at the end of the day, most developers pander to where the money is.
And whatever mess their tools leave behind falls back on businesses that rush to adopt without establishing a verifiable trust benchmark.
This trust verification has to happen first and not after things go sour and that’s part of what we help our clients do at Striv AI.
It's the first step in lowering AI governance complexity, as it helps unravel what you're deploying before the board gets to ask.
#AIGovernance #OpenClaw #AIGovernance #AgenticAI #AISecurity
@Omo_Adesawe@FinPlanKaluAja1 That’s where encryption in transit comes in. You can encrypt the data as it travels from that remote village to the central collation server in Abuja where it can be decrypted. It’s that simple
Over the past few weeks at Striv AI, I’ve been working on questions bordering on trust, governance, and security risks in agentic systems.
Most days involve tracing how models behave/might behave across workflows, data pipelines, and user interactions.
That work shows how technical design choices raise tough questions for governance, risk and compliance.
One area I’ve enjoyed is helping build internal artefacts that translate trust into measurable system indicators.
This includes mapping AI safety signals to ISO 42001, NIST AI RMF and EU AI Act.
It’s a useful way to turn abstract governance principles into practical checks and controls.
I’m happy to be working alongside our Chief Science Officer, C. Emre Koksal, Founder Jared Worley and a group of highly exceptional AI engineers.
They’ve sharpened my understanding of how much care quality systems demand.
Still learning.
Still helping to build the Trust Layer for safe interoperability of agentic AI systems.
#AIGovernance #AISecurity #ResponsibleAI #TrustInfrastructure #AgenticAI
A lot of companies will likely rush to adopt AI agents this year without a solid plan to check if things are still working as intended over time. It’ll be like launching a ship into open water and throwing the compass away. Which means, you might start on course, but you won't stay there.
Gartner Market Guide for AI TRiSM, notes that in 2026, 80% of unauthorized AI transactions will be caused by internal violations of enterprise policies rather than external malicious attacks.
One way to keep AI systems operating in line with approved governance policies is to commit both human and financial resources to AI performance evaluation (PE). Both matter.
Too often, organisations adopt automated PE tools without allocating sufficient budget or capacity for human-in-the-loop oversight. They forget that using AI to monitor AI still leaves gaps in accountability. When a secondary model evaluates a primary one, you risk a recursive loop where one black box judges another.
For organisations trying to play by the ISO/IEC 42001:2023 rule, clause 9.1 made reference to this issue. It mandates that organizations shall not just determine what to monitor and measure, but how to ensure measurement effectiveness. This is where the human element becomes crucial within the PE operation.
If your organisation is readying things to operationalise ISO/IEC 42001, clause 9.1, your AI governance actions should include:
• Standardize system monitoring and maintenance: document how your AI stays running through continuous performance evaluation and scheduled updates. This helps you apply control A.6.2.6.
• Establish clear metrics like the F1 value: use this to detect when a system deviates from benchmarks.
• Tailor technical documentation for stakeholders: determine exactly what information users, partners, and regulators need to see.
• Automate the recording of AI event logs: use solutions that will not only keep records of events but offer a more granular trust layer that provide immutable audit trails and full traceability of system events, decisions, and outcomes. This means logs that show who or what acted, under which policy constraints, on which task, with what data access, and whether behaviour stayed within expected bounds.
#AI #AIGovernance #ISO42001 #AIRiskManagement #ResponsibleAI #CyberSecurity
I’m excited to share that I’m starting a new position as AI Security and Governance Engineer @ https://t.co/WhSco6rnIp!
I’ll be working alongside the team to build and strengthen AI governance, trust, and security capabilities across the platform. This includes supporting governance frameworks for autonomous AI agents, helping integrate emerging regulatory requirements, supporting usability testing, and contributing to shaping client and partner messaging.
This new milestone is shaped by many years of intentional self-investment, continuous learning, and saying yes to opportunities that felt uncomfortable at the time. To anyone early in their journey, especially those coming from non-traditional backgrounds, progress compounds quietly before it shows. Stay curious, build real skills, and do not underestimate where consistency can take you.
#AISecurity #AIGovernance #trustinfrastructure #EnterpriseSolutions #ResponsibleAI
For those that like to deceive themselves I pity you.
Look at the discrepancy between what NASS CTC released and the gazetted copy
For those that support lies, continue! The truth is coming out
Hahahah!
I spent time reviewing the newly released NIST Cybersecurity Framework 2.0 Profile for Artificial Intelligence this weekend, and can say its timing makes sense given where the industry is headed. AI is increasingly moving from pilot tools into production, but with lots of uncertainties around ways to identify, measure, and govern the risks that come with it.
What this preliminary profile does is bring AI into cybersecurity territory by aligning it with NIST's CSF six core cybersecurity functions. That way, it gives organisations a familiar but practical reference point for weaving AI risk into existing security and enterprise risk management programs.
It also acknowledges AI’s growing role in day-to-day defensive operations, while factoring in how those same techniques are being leveraged by attackers at scale. The profile also assumes organisations are at different stages of AI adoption, and in doing so, increases its relevance regardless of maturity.
#AI #AIgovernance #Informationsecurity #AIsecurity