{"id":134181,"date":"2025-10-23T04:06:12","date_gmt":"2025-10-23T03:06:12","guid":{"rendered":"https:\/\/www.intelligentcio.com\/me\/?p=134181"},"modified":"2025-10-23T04:06:13","modified_gmt":"2025-10-23T03:06:13","slug":"inside-the-grid-inceptions-ashish-koshy-on-building-useful-ai","status":"publish","type":"post","link":"https:\/\/www.intelligentcio.com\/me\/2025\/10\/23\/inside-the-grid-inceptions-ashish-koshy-on-building-useful-ai\/","title":{"rendered":"Inside the grid \u2013 Inception\u2019s Ashish Koshy on building useful AI"},"content":{"rendered":"\n

Ashish Koshy talks about Artificial Intelligence as if it were a utility \u2013 not a spectacle but a grid. \u201cAt the base you have Khazna, then Core42 for GPU-as-a-service,\u201d he said. \u201cOn top is Inception \u2013 the model-building and product company.\u201d The stack matters because it explains how the Abu Dhabi group G42 tries to turn computation into outcomes without the usual detours. \u201cWe work hand in glove with domain companies \u2013 energy, healthcare, weather \u2013 so use cases are grounded in reality, not just lab demos.\u201d<\/p>\n\n\n\n

Inception began life in 2017 as a research institute for frontier models. That phase produced Jais, a large language model with an Emirati dialect spine, and cousins for Hindi and Kazakh, with Swahili and Azerbaijani under way. In 2024 the group pivoted. \u201cWe moved from \u2018Inception 1\u2019 \u2013 research first \u2013 to \u2018Inception 2\u2019,\u201d Koshy said. \u201cSame scientific depth but aimed at market-needed applications.\u201d He became chief executive late that year, though he had been with G42 since its earliest days. The move reads less like a handover than a narrowing of the beam.<\/p>\n\n\n\n

Local by design<\/strong><\/p>\n\n\n\n

The regional tilt was not a retrofit. Years before the present hype the UAE appointed a minister for AI and built MBZUAI, a university dedicated to the field. That foresight set a tone. \u201cAI must be democratised,\u201d Koshy said. \u201cNations should not be limited to models trained elsewhere with assumptions that do not fit local culture or policy.\u201d<\/p>\n\n\n\n

For him language is not a cosmetic layer but a vessel of norms. \u201cJais carries a strong Emirati flavour,\u201d he said. \u201cWe are doing the same for other languages because accuracy and tone in the local language affect trust and adoption.\u201d<\/p>\n\n\n\n

The practical stakes are obvious enough. Arabic content engines that understand idiom. Hindi-first customer service at scale. Enterprise tools that respond to how people actually speak rather than how documentation imagines they should. The point, Koshy said, is utility over theatre \u2013 a model that reads you as you are.<\/p>\n\n\n

\n
\"\"
Ashish Koshy, Chief Executive Officer, Inception, a G42 Company<\/figcaption><\/figure><\/div>\n\n\n

Beyond the demo<\/strong><\/p>\n\n\n\n

Koshy is wary of AI that never leaves the slide deck. \u201cDurable ROI comes when AI is embedded in core operations \u2013 not when it sits in a demo,\u201d he said. His team ships an In-Business suite across procurement, productivity and process improvement, including no-code tools for building agents. But he resists the idea that agents are plug-and-play.<\/p>\n\n\n\n

\u201cA procurement agent must reflect your delegation of authority, your policies, your ERP integrations,\u201d he said. \u201cYou need subject-matter experts in the loop. We co-design, set KPIs with the business, then measure against them.\u201d<\/p>\n\n\n\n

Some paybacks are quick. He points to a \u201cboard observer\u201d agent that acts as a companion in board meetings \u2013 compressing long decks into briefings, surfacing risks and action items so limited time can be spent on strategy rather than page-turning. Others take quarters. \u201cEnd-to-end procurement or investment agents span the value chain,\u201d he said.<\/p>\n\n\n\n

\u201cThey need robust integration and change management.\u201d The discipline is case by case: run proofs that touch the real workflow, scale only when both the technical and cultural bars are cleared.<\/p>\n\n\n\n

Partnerships keep the machine from trying to build every cog in-house. In energy, Inception leans on AIQ. For procurement, Fairmarkit. For model work, universities. \u201cDo not reinvent what you can partner on,\u201d Koshy said. \u201cInvest where you can differentiate.\u201d<\/p>\n\n\n\n

Data and the border<\/strong><\/p>\n\n\n\n

If models are the surface, data is the bedrock. In a region where sovereignty and privacy rules vary, Koshy begins with governance. \u201cWe deploy to the posture the client chooses \u2013 air-gapped on-prem, sovereign cloud or public cloud with client-held keys,\u201d he said.<\/p>\n\n\n\n

\u201cWe ship models into the client\u2019s tenant so data stays under their controls.\u201d For proofs without access to sensitive records, his teams generate synthetic datasets. Cross-border rules are \u201cdesign inputs, not obstacles,\u201d he said. The non-negotiables are simple: transparency and auditability.<\/p>\n\n\n\n

Responsible AI shows up in the plumbing rather than the press release. Group-level governance playbooks arrive with the product. Teams annotate trustworthy data, design auditable processes and implement risk controls. \u201cDecisions must be explainable and repeatable \u2013 not hallucinated,\u201d Koshy said. The bias is to publish when possible, even down to training datasets, because traceability invites scrutiny and hardens practice.<\/p>\n\n\n\n

Two fronts, one platform<\/strong><\/p>\n\n\n\n

Inception\u2019s market stance is tidy. \u201cWe run two fronts on one platform,\u201d Koshy said. The first is Corporate AI: back-office and knowledge-work functions like procurement, finance, legal and HR, wired with agent workflows customised per enterprise. The board observer sits here, licensed alongside the base platform so organisations can widen use over time.<\/p>\n\n\n\n

The second is domain-specific \u2013 stacks like Investment AI, fine-tuned on sector data to support sourcing, diligence and even voting workflows. The company is industry-agnostic yet enterprise-focused, which lets clients mix the horizontal with the vertical where impact is clearest.<\/p>\n\n\n\n

Ambition is measured in returns rather than slogans. \u201cIn three years I want \u2018AI in the region\u2019 to be synonymous with G42 and Inception because customers can point to measurable ROI,\u201d Koshy said. The plan involves a team that can handle national-scale problems, plus partnerships for when demand outpaces current capacity. He also wants early adopters to become messengers. \u201cLeaders who took a leap of faith with us should have the metrics to show it,\u201d he said.<\/p>\n\n\n\n

The human stays<\/strong><\/p>\n\n\n\n

For all the talk of agents, Koshy is blunt about the boundary. \u201cSkynet is great cinema, not a roadmap,\u201d he said. \u201cWe are in narrow AI with fast progress in physical AI, but a human remains in the loop in our systems.\u201d High-stakes domains demand proportionality: a missed calendar entry is tolerable, moving money is not.<\/p>\n\n\n\n

The human sets thresholds, supervises exceptions, owns the trade-offs. \u201cAI should move people from the mundane to the strategic,\u201d he said. The posture extends to culture. Talent joins for pay, environment and meaningful problems. Inception sells the third. \u201cWork on oil and gas optimisation one day, investment workflows the next, frontier models the day after,\u201d Koshy said.<\/p>\n\n\n\n

He looks for skill and attitude \u2013 people who like hard problems and accept the pace. The result is a shop where a 1 a.m. call happens because the challenge matters, not because anyone is forced.<\/p>\n\n\n\n

A builder\u2019s habit<\/strong><\/p>\n\n\n\n

Koshy\u2019s answers circle back to the idea of impact. Before Inception he helped build G42\u2019s healthcare arm from employee one through the peak of COVID, including the Emirati Genome Programme that many thought impossible at that scale. \u201cWe turned a lab into what I call the United Nations of sequencing,\u201d he said, \u201cby bringing together partners who rarely collaborate.\u201d AI is the next frontier \u2013 an \u201copen lab\u201d of compute, data and permission to build \u2013 and he sounds happiest when the work moves from pitch to production. \u201cI want us to leave deployed systems that matter,\u201d he said.<\/p>\n\n\n\n

The grid metaphor holds. A base of compute. A layer of models. A set of applications that live or die by whether they change the working day. Local language as a design choice, not a gloss. Governance as engineering, not theatre. The human \u2013 still, stubbornly \u2013 in the loop. \u201cDurable ROI is when AI becomes part of the operation,\u201d Koshy said. He made it sound less like a revolution than a craft \u2013 a set of fits, tests and integrations that, if you are patient, turn into outcomes.<\/p>\n","protected":false},"excerpt":{"rendered":"

Ashish Koshy talks about Artificial Intelligence as if it were a utility \u2013 not a spectacle but a grid. \u201cAt the base you have Khazna, then Core42 for GPU-as-a-service,\u201d he said. \u201cOn top is Inception \u2013 the model-building and product company.\u201d The stack matters because it explains how the Abu Dhabi group G42 tries to […]<\/p>\n","protected":false},"author":4338,"featured_media":126160,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[18762,20273,20272],"tags":[3211,908,21093,562,184],"class_list":["post-134181","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai","category-c-suite","category-technology","tag-ai","tag-digital-transformation","tag-inception","tag-middle-east","tag-uae"],"acf":[],"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/posts\/134181","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/users\/4338"}],"replies":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/comments?post=134181"}],"version-history":[{"count":1,"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/posts\/134181\/revisions"}],"predecessor-version":[{"id":134189,"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/posts\/134181\/revisions\/134189"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/media\/126160"}],"wp:attachment":[{"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/media?parent=134181"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/categories?post=134181"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.intelligentcio.com\/me\/wp-json\/wp\/v2\/tags?post=134181"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}