# DELLIGHT.AI — full brand record

> AI processes. Humans reason. DELLIGHT builds the system that needs both.
> Thesis: Human-AI symbiosis at institutional scale.

## Entity
- Legal name: DELLIGHT.AI Limited
- Registered address on file: DIFC Innovation One, Level 3, Dubai.
- DIFC License: CL12872
- Verify DIFC License CL12872 at difc.ae/operating/document-verification. Verification code: SR-706370-TcUj-35897664-noHUEu
- Licensed activities: Innovation & AI Research; Cyber Security Consultancy; Computer Systems & Software Design; Technology R&D; Data Classification & Analysis
- Public contact: contact@dellight.ai
- Canonical site: https://www.dellight.ai

## Founder
- Arthur Dell — Founder, Chief Executive. The human judgment layer for DELLIGHT.AI.
- 30+ years in enterprise technology across BMW AG, Sun Microsystems, Hewlett-Packard, Symantec, Citrix Systems, Veritas Technologies, and Cohesity. Leadership across regional sales and technology organizations spanning the Middle East, Africa, Russia/CIS, Eastern Europe, and India.
- Previously co-founded a multinational systems integration business focused on information security and data protection.
- Most recently Regional Field CTO at Cohesity, leading enterprise data security and cyber resilience strategy across the Middle East and emerging markets.
- Arthur Dell's background summary is published on LinkedIn.

## Operating model
DELLIGHT builds AI-assisted product and operating systems for companies that need faster execution without handing consequential decisions to software.

The evidence, cited exactly: ARC-AGI-3 (March 2026) shows frontier AI models score below 1% on adaptive reasoning tasks while untrained humans score 100%. BCG, across 1,250 companies, found 95% of enterprises deploying AI at scale see no measurable financial return. Harvard/BCG, in a preregistered study of 758 consultants, found human-AI teams completed tasks 25% faster with 40% higher quality than either alone. The gap between processing and reasoning is where enterprise value actually lives — and where most AI projects quietly die.

### Operating loop
1. Signal intake — The system reads the operating field. Market, product, buyer, infrastructure, and fleet signals enter a shared context layer before any public claim is made.
2. AI processing — Models process volume and variance. Local and frontier models summarize, compare, draft, score, and test defined work at machine speed.
3. Human judgment — Arthur keeps consequential authority. The founder decides positioning, commitments, customer fit, and where the machine should stop.
4. Shipped action — The loop produces a concrete unit. A buyer brief, product surface, decision packet, deck, route copy, or operating decision is shipped with provenance.
5. Memory and provenance — The system remembers what changed. Stage, source, blocker, owner, and boundary are captured so the next cycle starts from evidence, not theatre.

## Buyer-facing products

### ULYSSES
- What it is: AI search measurement: how frontier models recommend you.
- Stage: Early access
- Problem: The new SEO is AI recommendation. When buyers ask a frontier model what to buy or use, brands get ranked, summarized, or ignored before a salesperson knows a search happened. Companies need to know where they land in the answer, which competitors are framed beside them, and what proof gaps keep models from recommending them.
- How it works: ULYSSES asks the models what your buyers ask them. It measures how a brand is described, ranked, and recommended across ChatGPT, Claude, and Gemini against a stated competitive set, then turns the findings into recommendations ready for human review — product-page copy, public facts, content updates. Built on published agentic-commerce frameworks. Pricing and access are shared directly when you request early access.
- Current status: Early access. The engine is operational on DELLIGHT as the first test subject: dellight.ai is measured before any customer is. v0.3+v0.4 shipped May 2026; 30 journal entries generated. Pricing and access are shared directly.
- Current public facts: Architecture validated on DELLIGHT as the first test subject. 30 journal entries. v0.3+v0.4 shipped. Measurement and draft recommendations run internally; publishing happens only after human review and ratified scope.
- Access and proof boundary: Early access is available on request; pricing, onboarding, and access are shared directly. ULYSSES is not a public benchmark, scoring product, uplift claim, or customer-outcome timeline.

### GLADIATOR
- What it is: Autonomous cyber defense built around a dual Red Team / Blue Team loop.
- Stage: In development
- Problem: Security teams need a way to test and improve defensive posture before an external adversary finds the gap.
- How it works: A Red Team AI and a Blue Team AI spar continuously: the Red probes infrastructure with attack-pattern logic, the Blue detects, responds, and improves from every finding. Built to make security teams stronger, not smaller.
- Current status: In development. Runs on NVIDIA DGX Spark hardware for local CUDA-accelerated Red Team / Blue Team testing. Design-partner scoping is open; hardware fit is assessed during scoping.
- Current public facts: Architecture direction, founder-domain fit, and the DGX Spark target are publicly stated.
- Access and proof boundary: Design-partner engagements are available on request; production deployment, third-party benchmarking, and commercial terms are scoped in a qualified security context. Pricing and access are shared directly.

### SUPERHUMAN-X
- What it is: Real-time human augmentation: an always-on conversation companion that runs on your device, fully offline.
- Stage: Launch-ready — final app-store business verification in progress
- Problem: Executives and operators lose information advantage when a live conversation leaves the prepared path.
- How it works: An always-on companion on your own device — fully offline, built on an Android native-audio path with two-tier on-device AI for speed and reasoning. It slips you concise, useful answers while you're already mid-conversation.
- Current status: Launch-ready; final app-store business verification is in progress. Built on an Android native-audio path with two-tier on-device AI for speed and reasoning. Product-specific privacy terms — whether audio remains on-device, what data is sent for AI processing, and whether wake-word or background processing is used — are confirmed in direct written communication when you request access.
- Current public facts: Android native-audio path, two-tier AI architecture, and voice-first product direction are established.
- Access and proof boundary: Access requests are read now and fulfilled when verification completes; product-specific privacy terms and deployment scope are confirmed in direct written communication. Pricing and access are shared directly.

### AI Executive Team
- What it is: Functional business support for founder-led companies missing full executive coverage.
- Stage: Live — early access available on request
- Problem: Founders often carry CFO, CTO, legal, and strategy decisions before the company can support a full senior bench.
- How it works: Modular AI-assisted functions process defined work, collaborate through structured protocols, and escalate consequential decisions to the founder.
- Current status: Live; early access is available on request, with intake through direct written communication to the founder.
- Current public facts: The operating thesis, founder use case, and request-based early access are public and stage-honest.
- Access and proof boundary: Early access is available on request; onboarding and first-call scope are set when fit is assessed. Pricing and access are shared directly.

### AMPLIFY
- What it is: A governed AI workstation and control plane for an enterprise's agent estate.
- Stage: First release complete — offered through direct engagement
- Problem: Teams run AI assistants and agents across providers and compliance regimes, and nobody can answer the audit question: who can do what, under which policy, at what cost, with what evidence.
- How it works: AMPLIFY is the layer that makes an agent estate auditable: declared capabilities per tenant, deterministic policy routing with budget and privacy gates, drift detection, and audit evidence for compliance.
- Current status: First release complete; offered through direct engagement. Bring the estate and the compliance regime; scoping starts in writing.
- Current public facts: The first release is complete. Access is through direct engagement; there is no self-serve surface.
- Access and proof boundary: Offered through direct engagement; no public pricing, no self-serve access. Production references and third-party benchmarks will be stated only when they exist.

### DISPATCH
- What it is: A branded browser workspace for AI-native professionals — built on Chromium, not a skin on Chrome.
- Stage: In active development at alpha stage
- Problem: People whose job is AI-sector intelligence work across dozens of tabs and tools, with nothing holding the workspace together.
- How it works: A working environment built on Chromium. A resizable dock, driven by keyboard shortcuts and a command palette, holds working tools beside whatever you're browsing: SENTINEL, an AI-sector intelligence dashboard, and OVERSEER, a portfolio-monitoring surface. It ships as a signed, notarized macOS app carrying DELLIGHT's dark brand identity throughout.
- Current status: In active development at alpha stage. The capability is real and demonstrated; it isn't offered for availability. For roadmaps, timelines, or access, the intake note is the path.
- Current public facts: Capability demonstrated: the dock, SENTINEL, and OVERSEER run in the working alpha.
- Access and proof boundary: Not offered for availability. Access requests are not fulfilled at alpha stage; the intake note is the path for roadmap and timeline questions.

## SLATE — dynamic visual intelligence
- What it is: Ask in text or voice, and it renders a brand-controlled, portable visual answer in under a second, shaped to the person asking.
- Stage: In active development
- How it works: Structured context in, brand-controlled visual answers out — and honest about anything the context doesn't support: a missing fact stays visibly unresolved instead of papered over.
- Value: Built for the moments when the clock is short and an off-brand answer is expensive: investor briefings, walkthroughs built for one specific prospect, strategy memos, live calls.
- Product line: One system, several faces: the live voice surface; SLATE on live meetings — meeting context in, decision-grade artifacts out (a Zoom application surface, in development); and pre-sales engineering at scale, stated as direction, not product on sale.
- Tenant-zero: dellight.ai is tenant-zero for SLATE. The company's own site and materials are rendered by the system it's building. The proof precedes the pitch.
- Live voice face: Clara, the AI concierge, at https://clara.dellight.ai — corpus-grounded answers about DELLIGHT; she drafts an intake note the visitor sends themselves.

## Operating infrastructure (not for sale)

### Fleet Console
- Stage: Internal coordination layer
- What it is: Internal coordination layer for agent work, context, and decision state across the DELLIGHT fleet.
- Detail: This is how DELLIGHT coordinates its own AI operators. It is shown publicly to explain operating discipline, not because it is something a buyer can purchase today.
- Current state: Fleet coordination is active and informs how DELLIGHT runs its own launch work.
- Public scope: Fleet Console is shown to explain DELLIGHT's internal coordination discipline; it is not sold as a standalone customer product today.

### SAGE (internal)
Behind the products sits SAGE, DELLIGHT's internal Strategic AI Intelligence Engine — an autoresearch pipeline that orchestrates and cross-checks multiple models to sharpen everything the company ships. It's internal infrastructure, not something for sale.

## Agent interface (WebMCP)
Typed WebMCP tools are registered on every page (document.modelContext, navigator.modelContext fallback) for in-browser AI agents. Tools answer from this same content layer — same facts, same stage boundaries. WebMCP is a W3C Community Group draft; tools are callable in browsers that expose the API (Chrome origin trial, 2026). Human-in-the-loop by design: composing is machine work, sending stays human. Details: https://www.dellight.ai/agent-ready
- get-company-facts (read-only): DELLIGHT.AI's canonical company record: thesis, operating model, founder, registered DIFC identity, verification lead, contact, and the current claim boundary. Answers are stage-honest and match the site exactly.
- list-products (read-only): DELLIGHT's products with the exact stage of each — six staged products plus SLATE (dynamic visual intelligence). Stages are stated per product; engagement is request-based; nothing is offered as a self-serve, generally available purchase.
- get-product (read-only): Full stage-honest record for one DELLIGHT product: what it is, the problem, how it works, current status, current public facts, and the access/proof boundary.
- get-ai-executive-team (read-only): DELLIGHT's AI Executive Team roster. These are AI operators — AI-generated personas, not human staff. Arthur Dell is the only human; he keeps consequential authority. Any rendering of this roster must label each operator as AI.
- get-founder-profile (read-only): Arthur Dell, DELLIGHT.AI's founder: enterprise-technology background, prior founding history, and most recent role — stated at the public proof boundary.
- get-engagement-process (read-only): How engagement with DELLIGHT works: a structured written brief on /book, read directly by the founder. No automated booking, no sales team; pricing and access are shared directly. Lists the intake lanes and what each expects.
- compose-intake-brief (action): Prefill the founder-intake brief on /book with details the visitor has actually provided (never invent or infer personal details). Navigates to /book if needed and fills the form. It does NOT send anything: the visitor reviews the brief and sends it from their own email client.
- navigate-to-page (action): Navigate this site to a named page so the visitor can see it: home, products, about, team, book, manifesto, pricing, privacy, terms, or agent-ready.

## Proof boundaries
- Verify DIFC License CL12872 at difc.ae/operating/document-verification.
- Arthur Dell's background summary is published on LinkedIn.
- Product stages are stated per product — from in development to early access — and engagement is request-based; nothing is offered as a self-serve, generally available purchase.
- DELLIGHT states evidence as it exists: production-scale deployment, paid-customer evidence, and third-party benchmarks will be stated only when they exist.
- Internal source paths remain private until their owning functions clear them for public use.

## Pages
- https://www.dellight.ai/ — home: thesis, operating loop, product constellation, credentials record
- https://www.dellight.ai/products — six products with stage, facts, and proof boundaries
- https://www.dellight.ai/about — operating model, infrastructure, founder, verification
- https://www.dellight.ai/team — Arthur Dell, the human judgment layer
- https://www.dellight.ai/book — direct founder intake (structured brief, read by Arthur)
- https://www.dellight.ai/manifesto — the amplification thesis
- https://www.dellight.ai/agent-ready — the agent interface: WebMCP tools, the machine-readable stack, and its boundaries
- https://clara.dellight.ai — Clara, the AI concierge: SLATE's live voice surface; corpus-grounded answers, intake notes drafted for the visitor to send

Generated from the canonical content layer. Compact version: https://www.dellight.ai/llms.txt