AI CAPABILITY • LANDSCAPE
AI Landscape Navigator
The landscape is exploding. The gap between movers and waiters is widening.
More capability ships quarterly than used to ship in years. New models, tools, and platforms appear weekly. Competitive advantage is materialising now – not “someday.”
This living reference helps you understand what's out there, what matters, and where we stand. Not exhaustive – navigational.
A Living Reference
This page is updated as the landscape evolves. It reflects our current understanding and experience, not comprehensive market research. We include tools we've used, evaluated, or tracked closely. Last updated: December 2025.
Why Timing Matters
The landscape isn't just changing – the pace of change is accelerating.
Clock Speed Reality
Features ship faster than conferences can announce them. More capability is shipping quarterly than organisations used to deliver in 5-6 years of traditional technology change.
Leaders Pulling Ahead
The gap between organisations that “get” AI and those still experimenting is widening. Not because technology is inaccessible – everyone has access now – but because execution speed is separating leaders from laggards.
Model Commoditisation
The models themselves are increasingly commoditised. Your competitive advantage isn't which model you use – it's how quickly you build capability around it. Context, orchestration, and fluency matter more than model selection.
Ecosystem Thinking
It's not just about picking tools – it's about understanding vendor roadmaps, integration standards, and how pieces fit together. Technology decisions should be based on where providers are heading in 18-24 months, not just current features.
The Landscape at a Glance
ANTHROPIC
Claude models
Anthropic leads in safety and capability. Claude models excel at nuanced reasoning, coding, and following complex instructions. Their "Constitutional AI" approach prioritises helpfulness, harmlessness, and honesty.
Examples: Claude 4.5 Opus, Claude 4.5 Sonnet, Claude 4 Haiku
Strengths:
- +Exceptional at complex, nuanced tasks
- +Strong coding and technical capability
- +Excellent instruction following
- +Safety-conscious design
Considerations:
- •Smaller ecosystem than OpenAI
- •API access required for most use cases
- •Newer to consumer market
Our view: Our primary choice for complex work. We use Claude daily across all capabilities.
OPENAI
GPT models & ChatGPT
OpenAI pioneered the current AI era. ChatGPT remains the most widely known AI interface. Their models are capable across a broad range of tasks, with extensive ecosystem and integrations.
Examples: GPT-4o, GPT-4 Turbo, o1-preview, ChatGPT Plus
Strengths:
- +Largest ecosystem and integrations
- +Strong brand recognition
- +Broad capability across tasks
- +Extensive plugin/GPT store
Considerations:
- •Can be verbose or "assistant-like"
- •Rapid organisational changes
- •Heavy investment in consumer market
Our view: Solid all-rounder. Many clients already use ChatGPT; we help them get more value from it.
Gemini models
Google brings deep AI research heritage and integration with Google ecosystem. Gemini models are multimodal from the ground up, with strong reasoning and long context capabilities.
Examples: Gemini 2.0 Ultra, Gemini Pro, Gemini Flash
Strengths:
- +Native multimodality (text, image, video)
- +Google ecosystem integration
- +Very long context windows
- +Strong research foundation
Considerations:
- •Availability varies by region
- •Enterprise features still maturing
- •Less established in coding tasks
Our view: Strong option for multimodal and Google-integrated workflows.
XAI
Grok models
Elon Musk's AI venture, integrated with X (Twitter). Grok models are designed to be more direct and less filtered than competitors, with real-time access to X platform data.
Examples: Grok-2, Grok-1.5
Strengths:
- +Real-time information from X
- +Less guardrails on topics
- +Fast iteration pace
Considerations:
- •Limited enterprise features
- •Tied to X ecosystem
- •Early-stage organisation
Our view: Emerging option with unique X platform integration. Worth monitoring as capabilities develop.
META
Llama models (open-source)
Meta's open-source approach has democratised access to capable models. Llama can be run locally or on private infrastructure, offering control and privacy that hosted APIs cannot.
Examples: Llama 3.2, Llama 3.1, Code Llama
Strengths:
- +Open source and customisable
- +Can run locally/privately
- +No per-token API costs
- +Growing ecosystem
Considerations:
- •Requires technical expertise to deploy
- •Smaller models than frontier APIs
- •Self-managed infrastructure
Our view: Important for privacy-sensitive deployments and organisations with technical capability.
MISTRAL
European AI models
European-founded AI company offering competitive models with strong performance-to-cost ratios. Open-weight models available for self-hosting, with API access for convenience.
Examples: Mistral Large, Mixtral 8x22B, Mistral Small
Strengths:
- +European data sovereignty option
- +Strong price/performance
- +Open-weight models available
- +Multilingual strength
Considerations:
- •Smaller ecosystem than US providers
- •Enterprise features still developing
Our view: Good option for European data sovereignty requirements.
MICROSOFT 365 COPILOT
Productivity AI for M365
Microsoft 365 Copilot embeds AI assistance across the productivity suite. It draws on your organisation's data via Microsoft Graph to draft documents, analyse spreadsheets, summarise emails, and assist in meetings. Microsoft's AI play for M365-centric organisations of any size. (See Microsoft Graph under Integration Standards for the API that powers this.)
Examples: Word, Excel, PowerPoint, Outlook, Teams integration
Strengths:
- +Deep M365 integration (Word, Excel, Teams, Outlook)
- +Access to organisational data via Graph
- +Enterprise security and compliance
- +Familiar interface for existing M365 users
Considerations:
- •Requires M365 + Copilot licence (adds cost)
- •Quality depends on organisation's data hygiene
- •Less capable than Claude/ChatGPT for complex reasoning
- •Locked to Microsoft ecosystem
Our view: The default choice for M365-centric organisations. Convenient but not best-in-class for complex work – we often see teams using Copilot for drafts and Claude for refinement.
GOOGLE WORKSPACE + GEMINI
Productivity AI for Google Workspace
Google has integrated Gemini across Workspace apps – drafting emails in Gmail, generating content in Docs, analysing data in Sheets, creating presentations in Slides. For organisations already in the Google ecosystem, it's the natural AI layer.
Examples: Gmail, Docs, Sheets, Slides, Meet integration
Strengths:
- +Native Workspace integration
- +Strong collaboration features
- +Competitive pricing vs M365 Copilot
- +Gemini's multimodal capabilities
Considerations:
- •Google ecosystem lock-in
- •Feature parity still catching up to M365 Copilot
- •Data privacy concerns for some organisations
- •Less established in enterprise than Microsoft
Our view: The natural choice for Google Workspace organisations. Rapidly improving – worth evaluating if you're already in the Google ecosystem.
CURSOR
AI-native code editor
Cursor is a VS Code fork built around AI assistance. It understands your entire codebase and can make multi-file changes. The leading tool for serious AI-assisted development.
Examples: Claude integration, multi-model support, codebase understanding
Strengths:
- +Full codebase context awareness
- +Multi-file editing capability
- +Claude and GPT-4 integration
- +Professional-grade editor
Considerations:
- •Requires coding knowledge
- •Subscription cost
- •Learning curve for full capability
Our view: Our daily driver for all development work. Transformative for coding productivity.
GITHUB COPILOT
AI pair programmer
GitHub Copilot integrates AI assistance into your existing IDE. Best known for code completion and suggestions, with expanding capabilities in chat and workspace features.
Examples: Code completion, chat, Copilot Workspace
Strengths:
- +Works in existing editors (VS Code, JetBrains)
- +Strong code completion
- +GitHub ecosystem integration
- +Enterprise features available
Considerations:
- •Less context-aware than Cursor
- •Primarily completion-focused
- •Subscription required
Our view: Good for teams already in GitHub ecosystem who want lighter integration.
WINDSURF
AI coding environment
Codeium's Windsurf editor offers similar capabilities to Cursor with a focus on speed and multi-model flexibility. Growing alternative in the AI-first editor space.
Examples: Multi-model support, context-aware editing
Strengths:
- +Fast and responsive
- +Multi-model support
- +Competitive pricing
- +Free tier available
Considerations:
- •Newer, less established
- •Smaller community
- •Feature parity still developing
Our view: Worth trying as Cursor alternative. Good for cost-conscious teams.
REPLIT
AI-powered browser IDE
Replit brings AI-assisted coding to the browser with instant deployment. No local setup required. Great for learning, prototyping, and teams without DevOps capability.
Examples: Ghostwriter, instant deployment, collaboration
Strengths:
- +No local setup needed
- +Instant deployment
- +Collaborative features
- +Great for learning
Considerations:
- •Less powerful than local editors
- •Running costs can scale
- •Limited offline capability
Our view: Excellent for rapid prototyping and teams without infrastructure expertise.
VS CODE
The extensible AI ecosystem
VS Code is the world's most popular code editor, and its extension ecosystem makes it a platform for AI development tools. Official Claude Code extension (Sept 2025) brings real-time AI assistance, checkpoints, and MCP integration directly into VS Code.
Examples: Claude Code extension, GitHub Copilot, Continue, Cody
Strengths:
- +Massive extension ecosystem
- +Official Claude Code integration
- +Familiar to most developers
- +Free and open source
Considerations:
- •Not AI-native like Cursor
- •Extensions vary in quality
- •Requires configuration for optimal AI use
Our view: The ecosystem play. If you prefer VS Code, Claude Code extension brings powerful AI assistance without switching editors.
GOOSE
Open-source autonomous AI agent
Goose is Block's open-source AI agent that goes beyond code suggestions. It autonomously installs dependencies, executes code, runs tests, and orchestrates workflows. MCP-native (Block co-developed MCP with Anthropic). Desktop app and CLI, works with any LLM, and lets you hot-swap models mid-conversation.
Examples: Autonomous coding, MCP-native, any LLM, Block engineering
Strengths:
- +Fully autonomous execution
- +Open source (Apache 2.0)
- +MCP-native integration
- +Any LLM, hot-swap models
Considerations:
- •Newer than established tools
- •Requires LLM API access
- •CLI-first (desktop app newer)
Our view: Exciting open-source option from an MCP co-creator. Worth watching as the agentic AI space matures.
GOOGLE ANTIGRAVITY
Agent-first IDE
Google's agent-first IDE, launching the next generation of development environments. Antigravity coordinates agents across editor, terminal, and browser surfaces, with task-based abstractions and multi-agent management from central control. Built on Gemini models.
Examples: Cross-surface agents, task-based monitoring, multi-agent orchestration
Strengths:
- +Cross-surface agent coordination (editor + terminal + browser)
- +Task-based abstractions for trust
- +Multi-agent management across workspaces
- +Google ecosystem integration
- +Browser-in-the-loop capabilities
Considerations:
- •Very new (launched Nov 2025)
- •Locked to Google/Gemini ecosystem
- •Agent-first may have learning curve
- •Limited track record vs established tools
Our view: Google's ambitious entry into the AI-first IDE space. Cross-surface coordination is genuinely differentiated - worth watching as it matures. Particularly interesting for teams already in Google ecosystem.
LOVABLE
AI app builder
Lovable generates full-stack applications from natural language descriptions. Built on React and Supabase, it can create functional MVPs rapidly with minimal coding.
Examples: Full-stack apps from prompts, React/Supabase stack
Strengths:
- +Full applications from prompts
- +Professional tech stack
- +Rapid prototyping
- +Iterative refinement
Considerations:
- •Limited customisation depth
- •Generated code needs review
- •Subscription costs
Our view: Powerful for rapid MVPs. Best combined with coding knowledge for refinement.
BOLT
AI web app generator
Bolt creates full-stack web applications entirely in the browser. Strong at generating working prototypes quickly, with ability to export code for further development.
Examples: Full-stack apps, instant preview, export capability
Strengths:
- +Browser-based, no setup
- +Fast prototype generation
- +Code export capability
- +Growing template library
Considerations:
- •Quality varies by complexity
- •Export code needs cleanup
- •Limited backend sophistication
Our view: Good for quick prototypes and validating ideas before investing in development.
V0 BY VERCEL
AI UI component generator
Vercel's v0 generates React components from natural language. Focused on UI rather than full applications, producing high-quality components using modern design systems.
Examples: React components from prompts, shadcn/ui integration
Strengths:
- +High-quality UI components
- +shadcn/ui integration
- +Production-ready output
- +Iterative refinement
Considerations:
- •UI-focused, not full apps
- •React ecosystem only
- •Requires integration work
Our view: Excellent for quickly generating UI components for existing projects.
SHAKESPEARE
Open-source AI development platform
Shakespeare is an open-source, browser-based AI development platform. Unlike Lovable or Bolt, you bring your own API keys and everything runs client-side. Built-in Git integration, deploy anywhere, no vendor lock-in.
Examples: Browser-based IDE, BYOK (bring your own keys), Git integration
Strengths:
- +Open source, no lock-in
- +Bring your own API keys
- +Client-side processing
- +Built-in Git integration
Considerations:
- •Requires own API keys
- •Less polished than commercial alternatives
- •Smaller community
Our view: Interesting open-source alternative for those who want control and transparency.
N8N
Open-source workflow automation
n8n offers visual workflow automation with the option to self-host. Extensive integrations including AI services. Fair-code model provides transparency while allowing commercial use.
Examples: Visual workflows, self-hostable, AI integrations
Strengths:
- +Self-hosting option
- +Visual workflow builder
- +Extensive integrations
- +Fair-code transparency
Considerations:
- •Learning curve for complex workflows
- •Self-hosting requires maintenance
- •Less polished than commercial alternatives
Our view: Our preferred automation platform. Self-hosting provides control and cost efficiency.
MAKE (INTEGROMAT)
Visual automation platform
Make (formerly Integromat) provides powerful visual automation with extensive pre-built integrations. More sophisticated than Zapier for complex workflows, with better value at scale.
Examples: Drag-and-drop workflows, 1500+ integrations
Strengths:
- +Visual workflow design
- +Excellent integrations library
- +Complex logic support
- +Good value at scale
Considerations:
- •Learning curve for power features
- •Operation-based pricing
- •No self-hosting option
Our view: Strong choice for teams wanting hosted automation without coding.
ZAPIER
No-code automation
Zapier pioneered no-code automation and remains the most accessible option. Best for simple, linear workflows. Extensive integrations make it easy to connect almost any service.
Examples: Simple triggers and actions, 6000+ integrations
Strengths:
- +Most accessible interface
- +Huge integration library
- +Quick setup
- +Well-documented
Considerations:
- •Limited complex logic
- •Expensive at scale
- •Simple workflows only
Our view: Good entry point for automation. Outgrow to n8n or Make for complex needs.
MCP (MODEL CONTEXT PROTOCOL)
Industry-standard AI integration protocol
MCP is the open standard for connecting AI to external tools and data. Originally created by Anthropic, it was donated to the Linux Foundation's Agentic AI Foundation (Dec 2025), co-founded with OpenAI and Block, and supported by Google, Microsoft, AWS, and others. Now adopted by ChatGPT, Cursor, Gemini, VS Code, and more.
Examples: Tool use, data sources, service connections
Strengths:
- +Industry-wide standard (Linux Foundation)
- +Massive adoption (10,000+ servers, 97M+ SDK downloads)
- +Supported by all major AI providers
- +Clean, portable integration pattern
Considerations:
- •Still evolving rapidly
- •Requires some technical setup
- •Best tooling in TypeScript/Python
Our view: The integration standard for AI. Industry backing makes this the safe, future-proof choice.
LANGCHAIN
LLM application framework
LangChain provides building blocks for LLM applications: chains of operations, agents with tools, retrieval-augmented generation, and conversation memory. The most popular framework for building AI applications.
Examples: Chains, agents, retrieval, memory
Strengths:
- +Comprehensive toolkit
- +Large community
- +Model-agnostic
- +Extensive documentation
Considerations:
- •Can add complexity
- •Abstractions may obscure behaviour
- •Fast-changing API
Our view: Useful for complex AI applications. Evaluate whether abstraction is worth the complexity.
OPENAI API
De facto industry standard
OpenAI's API has become the de facto standard that many providers emulate. Even non-OpenAI models often offer OpenAI-compatible endpoints, making it a common integration target.
Examples: Chat completions, embeddings, function calling
Strengths:
- +Industry standard format
- +Many compatible providers
- +Well-documented
- +Stable interface
Considerations:
- •OpenAI-centric design
- •May not expose all model features
- •Lock-in risk if using OpenAI-specific features
Our view: Useful standard for portability. Consider native APIs for model-specific features.
MICROSOFT GRAPH
M365 data and services API
Microsoft Graph is the unified API for accessing Microsoft 365 data and services. Available to any M365 subscriber (Business or Enterprise plans), it's the data layer that powers M365 Copilot – and can power your own AI applications. If your organisation uses M365, Graph is how AI gets context about your people, documents, and activities.
Examples: User data, emails, calendars, files, Teams, SharePoint
Strengths:
- +Single API for all M365 data
- +Available to SMEs (M365 Business plans)
- +Powers M365 Copilot under the hood
- +Rich organisational context
Considerations:
- •Microsoft ecosystem lock-in
- •Permissions model has learning curve
- •Requires some technical setup
- •Data quality affects AI outputs
Our view: If you're on M365, Graph is how you connect AI to your organisational data. Not just for enterprises – SMEs can use it too.
PERPLEXITY
AI-powered search
Perplexity combines LLM capability with real-time search. It provides cited, sourced answers to questions, making it valuable for research and fact-checking.
Examples: Research, citations, real-time information
Strengths:
- +Real-time information
- +Source citations
- +Research-focused
- +API available
Considerations:
- •Subscription for full features
- •Less suitable for creative tasks
- •Web-focused
Our view: Essential for research tasks requiring current information and sources.
NOTEBOOKLM
Document analysis by Google
Google's NotebookLM lets you upload documents and interact with them through AI. Excellent for synthesising research, understanding complex documents, and generating audio summaries.
Examples: Upload documents, ask questions, generate summaries
Strengths:
- +Document-grounded responses
- +Audio overview generation
- +Free tier available
- +Good for research synthesis
Considerations:
- •Google ecosystem
- •Limited export options
- •Document size limits
Our view: Excellent free tool for document analysis and research synthesis.
FIRECRAWL
Web scraping for AI
Firecrawl extracts clean, AI-ready content from websites. It handles the complexity of modern web pages, returning structured data suitable for LLM processing.
Examples: Clean web data extraction, API access
Strengths:
- +Clean data extraction
- +Handles dynamic content
- +API-based
- +LLM-optimised output
Considerations:
- •Usage-based pricing
- •Requires API integration
- •Some sites may block
Our view: Our preferred tool for web data extraction. Clean output saves significant processing.
GOOGLE AI STUDIO
Gemini playground & API access
Google AI Studio is the free interface for experimenting with Gemini models. Test prompts, compare model outputs, and generate API keys for integration. A good starting point for exploring Google's AI capabilities.
Examples: Model testing, prompt development, API key generation
Strengths:
- +Free access to Gemini models
- +Easy prompt experimentation
- +Direct API key generation
- +Multimodal testing (text, image, video)
Considerations:
- •Google account required
- •Less polished than ChatGPT interface
- •Some features region-restricted
Our view: Useful free tool for testing Gemini capabilities before committing to integration.
MAPLE AI
Privacy-first AI chat
Maple AI positions itself as "The Signal of AI" – end-to-end encrypted chat with various AI models including DeepSeek R1. All processing happens in secure enclaves with zero data retention. For professionals handling sensitive information.
Examples: End-to-end encrypted chat, multi-model access, zero data retention
Strengths:
- +End-to-end encryption
- +Zero data retention
- +Multi-model access
- +Desktop and mobile apps with sync
Considerations:
- •Subscription for full features
- •Newer, smaller user base
- •Bring-your-own-key for some models
Our view: Important option for privacy-sensitive use cases. Worth considering for confidential work.
OLLAMA
Run AI models locally
Ollama makes it easy to run open-source AI models on your own machine. Download and run Llama, Mistral, and many other models locally with simple commands. No API costs, complete privacy, offline capable.
Examples: Local Llama, Mistral, Code Llama, private deployment
Strengths:
- +Complete privacy – data never leaves your machine
- +No per-token costs
- +Offline capable
- +Simple setup and model management
Considerations:
- •Requires capable hardware (GPU recommended)
- •Smaller models than cloud APIs
- •Self-managed updates
Our view: Essential tool for local AI experimentation and privacy-sensitive deployments.
How We Navigate This
With so many options, how do you choose? Here's our approach.
Start with the Problem
Don't start with “what AI should we use?” Start with “what problem are we solving?” The tool follows from the task, not the other way around.
Favour Simplicity
The simplest tool that solves the problem is usually the right choice. Complexity has ongoing costs. Start simple; add sophistication when you hit limits.
Build for Portability
The landscape changes fast. Avoid deep lock-in where you can. Use standards (MCP, OpenAI-compatible APIs) that let you switch if better options emerge.
Test with Real Work
Demos impress; production reveals. Before committing, test tools on your actual tasks. What works in a demo may struggle with your specific context.
What's Not Here
Comprehensive Coverage
This isn't a complete market survey. We focus on tools we've used or seriously evaluated. Many good options aren't listed because we haven't worked with them.
Full Enterprise Stack
We cover M365 Copilot and Graph, but not the full enterprise AI stack (Copilot Studio, Power Platform AI, Salesforce Einstein, ServiceNow, etc.). These require enterprise-specific context.
Image/Video Generation
Midjourney, DALL-E, Sora, etc. are important but outside our primary focus. We concentrate on text-based AI for business operations.
Pricing Details
Pricing changes frequently. We mention pricing considerations but don't list specific prices. Check provider websites for current rates.
From Landscape to Practice
Understanding the landscape is step one. Making it work for your organisation is where we help.
Context Engineering
The right information at the right time. How to design systems that give AI what it needs.
Learn more →Agents & Orchestration
One agent, infinite expertise. Skills-based AI systems that compound value.
Learn more →AI Skills & Fluency
The bottleneck isn't tools – it's people and culture. Building genuine capability.
Learn more →Need Help Navigating?
The landscape is overwhelming. We've been navigating it daily. Let's talk about what makes sense for your situation.