Technology firm Lashan Digital has launched AISentri to help organisations give the visibility of AI usage such as which AI tools their staff are using, what vulnerabilities do they have, how much are they costing etc – and prove they are using them safely and responsibly
Which AI tools are your staff using? LASHAN Digital has launched AI Sentri to give organisations a clear answer.
The technology consultancy works with social housing providers on enterprise architecture, data, and AI. AISentri, which is now fully available, is its latest product.
Viswa Gogineni, Lashan’s founder and chief executive, said: “We are a consulting and engineering firm that does end-to-end services for our clients, right from architecture and strategy to delivery of those solutions through our engineering capability.”
The engineering is technological rather than physical, covering software, data, and AI. The firm also turns problems it sees repeatedly into products. “We also have very specific solutions for repeatable problems that we identify in client organisations,” Viswa said. “We package them as products and accelerators to avoid implementing the same thing again and again.”
Much of its work begins with enterprise architecture, which Viswa described as planning rather than reacting: “It’s not reacting to the problems but having a plan of action ahead of time, based on your business strategy for the next three years, five years, and what you want to achieve as a business.”
He added: “Sometimes you have strategies in every department, and enterprise architecture unifies all of them together and creates a unified roadmap for the whole organisation.”
Closing the gap between business and technology
Viswa said the challenges Lashan has seen in social housing, where it has worked for about five years now, were familiar from other sectors.
Medium-sized businesses elsewhere struggled with fragmented systems and the lack of a single source of truth, he said: “They go through the transformation, they go live, and the systems still don’t work as per the situation.”
The difference lay in scale and urgency, and in decision-making, which he said was generally quicker in the private sector.
He said the aim was to close the gap between what technology is meant to deliver and what departments need from it. “Technology is not a separate department,” he said. “They are the enabler for organisations to deliver services successfully.”
That includes curbing “shadow IT”, where departments buy their own tools, through governance forums in which business teams take part in technology decisions. “That way they have a say in what technology is trying to do, and they don’t go and buy off-the-shelf products on their own that don’t connect with the rest of the ecosystem,” Viswa said.
Long programmes add to the pressure. Transformations can run for three or four years, he said, with old and new systems operating in parallel. “It’s not like old legacy in one architecture, new in another architecture, everybody double keying everything,” he said. “That’s not sustainable.”
The VIVID programme
Lashan worked with housing association VIVID on a programme of that kind. Viswa said the work began with an architectural maturity assessment, followed by a target architecture, a roadmap, and input to the business case and vendor selection.
Lashan then produced solution blueprints for each phase and ran architectural governance through design authorities. The team was also responsible for defining and delivering the data migration approach for which an independent review by KPMG concluded that the overall migration strategy, governance, and execution approach were stronger than those typically observed across comparable programmes within the sector.
The team also built an integration hub to connect VIVID’s systems.
The foundations later extended to AI. Viswa said Lashan worked with VIVID to define its AI strategy and adoption roadmap. The AI strategy helped to identify the tools and additional skills required to deliver AI solutions.
Making AI visible
Kieran Keene, Lashan’s chief product officer, said AISentri addresses a visibility problem.
“A lot of businesses don’t really know what AI they’ve got running in their business,” he said. “One department has decided to launch a project doing X, another department is doing something else. They don’t talk to each other.”
The second aim was control. He said the tool helps organisations evidence compliance with the EU AI Act and GDPR. “If a regulator comes and says, ‘Why did you do that project? Why did you think it was a legally viable thing to do?’, they can just export it from the system and say, ‘Here you go. There’s our justification,’” he said.
AISentri is deliberately standalone, so it does not become another system needing its own oversight. “It’s not intended to link into all your other systems,” Kieran said. “It’s meant to be a governance layer that says, ‘We did this for this reason. This is why we did it. And this is the legal basis and the controls we have around it.’”
The list included “almost every large language model that is on the market”, he said, along with AI video generators used for user guides, and tools such as Grammarly that connect to documents.
VIVID can now log why each tool is used and what safeguards are in place against sensitive data being entered. Kieran said each tool is assessed on its own merits, with training and, in some cases, technical restrictions applied according to the risk.
Each AI initiative also receives two scores, he said. “One is your governance score, which shows you how risky the system is. And one is the return on investment that you receive from that system, from a financial point of view, or from a time point of view.”
Comparing the two helps organisations decide where to focus. “If you see this is a really poor scoring system, we’re going to have to put a lot of effort into governing it and it’s not giving us any returns, it gives you a whole different decision-making perspective,” Kieran said. “It’s a very quick view to say, ‘Actually, we don’t need it. It’s not worth putting the effort in.’”
Kieran said the aim was to help organisations use AI “intelligently and carefully”. “It’s all about protecting the customer data and protecting the business data from unintended consequences,” he said.
Lashan Digital is a Housing Digital Stakeholder and an architecture-led digital, data and AI transformation partner for the housing sector.
Main image: Viswa Gogineni, Founder and Chief Executive (left) and Kieran Keene, Chief Product Officer, Lashan Digital
Read next: Charities call for action as London rough sleeping rises
Are you a social housing professional? Sign up for a FREE MEMBERSHIP to upload news stories, post job vacancies, and connect with colleagues on our secure social feed.