slug: new-ai-tools-august-2027 title: New AI Tools — August 2027 Edition date: 2027-08-15 description: The best new AI tools launched in August 2027 — covering new releases, major updates, and standout tools across content creation, coding, productivity, and design.
Target keyword: new ai tools august 2027 | Last updated: Aug 2027
August 2027 delivered the Q3 update cycle that AI tool watchers had been tracking since June roadmap previews. Video AI made the jump from "impressive demo" to "production-grade" with Runway's Gen-4 launch, which resolves the subject-consistency problem that kept scripted content production out of reach. Cursor 2.0 shipped Background Agents — the feature that moves agentic coding from single-file autocomplete to genuine multi-file, multi-step development. Anthropic's Claude 5 reset the frontier reasoning benchmark. And Figma finally moved its AI features from eight months of limited beta to general availability, unblocking design teams that had been waiting since early 2027. Alongside the headliners, a quieter wave of focused tools shipped in August across research, meetings, and brand design that are worth attention even if they didn't make broader coverage. Here is what launched, what it costs, and whether it's worth adding to your stack.
What Happened in AI in August 2027
Three patterns defined the month:
Free Download
50 AI Tools by Category
The definitive guide to the AI tools actually worth using — curated by the team at DotProTools.
Download the free guide →You will also get our weekly Digest — unsubscribe from either anytime.
Agentic coding became a professional default. Cursor 2.0, updated GitHub Copilot Workspace support, and stability improvements to Replit Agent all shipped in August. Developers who had been using agentic coding tools for isolated or low-stakes tasks are now applying them to end-to-end feature builds and codebase migrations. The remaining friction is organizational — code review processes, security approvals for agent-authored changes — rather than product quality. The tools are ready; the workflows around them are catching up.
Frontier model competition intensified. Claude 5 arrived in August and directly challenged GPT-5's position on reasoning and long-context benchmarks. For any tool built on API access to frontier models — writing assistants, research platforms, analysis tools — access to a significantly more capable base model compounds forward: expect the downstream quality improvements to be visible across the AI tools ecosystem through Q4 2027.
Video AI crossed the production threshold. Runway Gen-4's "consistent character" capability — maintaining coherent subject identity across multi-scene video without manual consistency anchoring — is the specific technical feature that had kept video AI out of scripted production workflows. With Gen-4, that constraint is lifted for the majority of common production scenarios. Marketing teams that have been watching the video AI category from the sidelines have a real reason to act now.
Best New AI Tools in August 2027
| Tool | Category | What's New | Best For |
|---|---|---|---|
| Runway Gen-4 | AI video | Consistent character across scenes, 90s clips, Director Mode | Scripted brand content, agencies |
| Cursor 2.0 | AI coding | Background Agents, Project Memory, unified diff review | Full-stack dev teams |
| Figma AI (GA) | AI design | General availability, Design Generation from component library | Product design teams |
| Claude 5 | AI model | Extended reasoning, 500K context, Deep Analysis mode | Research, legal, complex analysis |
| Notion AI 3.0 | Productivity | Database Q&A, AI Automations, workspace-aware writing | Ops teams, project managers |
| Elicit 3.0 | Research AI | PRISMA review workflows, 50-paper extraction | Academic researchers |
| Luma Dream Machine 2.0 | AI video | 45s render time, improved consistency | Rapid video prototyping |
| Grain AI v3 | Meeting AI | Meeting theme clustering, pattern detection | Team managers |
| Recraft v5 | AI design | Brand consistency, vector generation | Brand and marketing assets |
| Writer Enterprise v3 | Content AI | House style enforcement at paragraph level | Enterprise content operations |
Deep Dives: Top New AI Tools This Month
Runway Gen-4
Runway's fourth-generation model is the video AI update the industry has been anticipating since Gen-3 hit its quality ceiling in late 2026. The headline feature is consistent character rendering across scenes — Gen-4 maintains coherent visual identity for subjects across cuts, camera angles, and environmental settings without requiring manual consistency anchoring between prompts. This was the single largest practical limitation in AI video for scripted content; Gen-4 resolves it for the majority of common production scenarios.
The quality improvements beyond consistency are also significant. Gen-4 supports clips up to 90 seconds (from 18 seconds in Gen-3), handles complex lighting transitions more naturally, and produces camera movement that reads as directed rather than physically approximate. The new Director Mode UI puts standard cinematography controls — shot type, camera movement, depth of field — in the primary interface rather than buried in prompt syntax. For users with film production backgrounds who found Gen-3's text-only prompt system awkward, this is a substantial improvement in day-to-day usability.
Pricing: Creator plan ($15/mo) includes 125 credits; Standard ($35/mo) includes 625 credits; Unlimited ($95/mo) covers unlimited standard-quality generation with priority queue for extended clips. Commercial licensing is included on all paid tiers.
One honest limitation at launch: the consistency feature works reliably for humanoid subjects but shows occasional drift with animals and heavily stylized visual treatments. Runway has acknowledged this and it is expected to improve in subsequent model updates.
Who it's for: Marketing agencies, brand content teams, and filmmakers producing scripted short-form content who previously hit the consistency wall with Gen-3 or Sora.
Learn more: Best AI Video Generators in 2027
Cursor 2.0
Cursor's second major version is the coding AI market's most credible demonstration that agentic development — where the AI handles multi-file, multi-step changes with minimal scaffolding — is ready for professional use. The 2.0 release centers on two features: Background Agents and Project Memory.
Background Agents run in a separate context from your active editing session. You describe a task ("add input validation to all API endpoints," "migrate the auth module from JWT to session tokens"), Cursor agents execute changes across the full codebase, and you review a unified diff when they complete — similar to reviewing a PR from a junior developer. The agents handle imports, cross-file consistency, and test updates for covered code paths.
Project Memory is the other major addition: Cursor now builds and maintains a persistent map of your codebase's architecture, conventions, and recurring patterns. Earlier versions required re-explaining your project structure at the start of each session; 2.0 loads that context automatically and references it in every suggestion and agent action. For teams with large, established codebases, this closes the quality gap between Cursor on greenfield projects versus legacy code.
Pricing: Hobby plan is free with limited agent actions; Pro is $20/mo (500 agent actions, priority model access); Business is $40/user/mo (unlimited agent actions, SSO, audit logs).
The main trade-off: Background Agents introduce a review step that slows flow on small, quick changes. The feature is opt-in and works best for medium-to-large-scope tasks — the kind of changes where the review step is proportionate to the scope.
Who it's for: Full-stack developers and software teams on established codebases who want agentic coding beyond autocomplete and single-file edits.
Learn more: Best AI Coding Tools in 2027
Figma AI (General Availability)
After eight months of limited beta, Figma AI shipped to all Professional and Organization plan subscribers in August. The GA release includes four primary features that were refined through the beta period: Design Generation, Layers Cleanup, Copy Refinement, and Prototype to Code.
Design Generation is the centerpiece: describe a UI component or screen layout in natural language, and Figma AI produces a design that draws from your team's existing component library rather than generic placeholder elements. Beta users consistently report that it correctly references saved components around 80% of the time — a significant acceleration over starting from scratch, with a manageable 20% requiring a manual component swap.
Layers Cleanup addresses a longstanding collaborative friction point: AI audits your file's layer structure, identifies inconsistent naming, ungrouped elements, and detached component instances, and proposes a cleanup you can apply in one click. For teams inheriting messy design files — or managing files across many contributors — this is the most immediately practical feature in the release.
Prototype to Code converts Figma frames to React or HTML/CSS. The output is cleaner than the 2026 beta but still requires developer review before production use. Treat it as a scaffold generator rather than a production-ready code exporter.
Pricing: Figma AI is included in the Professional plan ($15/user/mo) and Organization plan ($45/user/mo). Starter plan users get 25 Design Generations per month at no additional cost.
Who it's for: Product designers and design teams looking to accelerate ideation, reduce file maintenance overhead, and improve handoff quality to development.
Learn more: Best AI Design Tools in 2027
Claude 5
Anthropic shipped Claude 5 in August with three capability improvements that meaningfully separate it from Claude 4: significantly extended reasoning on complex multi-step problems, a 500K context window that handles dense technical documents more reliably than the previous 200K window, and a new Deep Analysis mode that runs extended chain-of-thought reasoning for complex tasks at the cost of slower response time.
The reasoning improvement is the most practically significant. Claude 5 scores above GPT-5 on standard complex reasoning benchmarks — the first time Anthropic has held the frontier reasoning position since Claude 3 Opus in early 2024. For users whose primary use cases involve multi-step analysis, legal document review, and code generation for complex systems, the 5.0 update is a genuine step change, not an incremental polish.
Deep Analysis mode is the most distinctive new interface feature. When enabled, Claude explicitly lays out its reasoning steps before delivering an answer — a chain-of-thought scratchpad that is visible rather than hidden. Users report this is most useful for debugging complex problems, where seeing Claude's reasoning path makes it easier to identify exactly where an analysis diverges from the expected outcome.
Claude 5 is available on Claude Pro ($25/mo) and Claude for Work (custom enterprise pricing). API access is available through Anthropic's console at approximately $12/million input tokens and $60/million output tokens for the full 5.0 model. A smaller, faster Claude 5 Haiku model for high-volume API use cases is priced significantly lower.
Who it's for: Researchers, analysts, legal professionals, and developers who hit the reasoning or context ceiling on Claude 4.
Notion AI 3.0
Notion shipped its 3.0 AI update in August, centered on database-aware AI — the feature the team previewed at their June developer event. The headline capability: Notion AI can now traverse linked databases, filter across properties, and synthesize cross-database insights. It moves from a writing aid that happened to live in Notion to an actual analytical layer on top of your workspace data.
Database Q&A is the most-used feature in early access: ask "which projects are behind schedule and owned by the marketing team?" and Notion AI queries your project database, applies the filter logic, and returns a natural-language summary alongside a linked filtered view. For teams that have invested in structuring their data in Notion, this is the unlock that makes that investment pay off analytically — not just organizationally.
AI Automations (in beta at launch) allow trigger-based workflows: "when a project status changes to At Risk, draft a summary and notify the project lead in Slack." Early access reports suggest reliability is high for simple trigger-to-action flows and requires iteration for multi-step sequences.
The 3.0 update also addressed a consistent complaint from Q2: Notion AI's writing suggestions were generic and not workspace-context-aware. The 3.0 model explicitly incorporates your Notion content as context for writing assistance, producing suggestions that reference your actual projects, team terminology, and documented decisions.
Pricing: Notion AI remains a $12/user/month add-on to any plan, included in Business and Enterprise. The full 3.0 feature set is available to all AI subscribers from August 15.
Who it's for: Operations teams, project managers, and knowledge workers who have centralized their work in Notion and want analytical intelligence on top of their workspace data.
Under-the-Radar Picks
Five August launches that didn't break through to broader coverage but are worth tracking:
Elicit 3.0 — The AI research assistant for academic literature shipped v3 with systematic review support: Elicit can generate a full PRISMA-format review structure, extract data across 50+ papers simultaneously, and flag contradictory findings across studies. Essential for researchers producing evidence reviews at volume. Free for individuals; Pro at $20/mo.
Luma Dream Machine 2.0 — Luma's video generation model updated with improved consistency and substantially faster generation (45-second render for a 5-second clip, down from 3+ minutes in v1). Still trails Runway Gen-4 on quality but is faster and cheaper for rapid iteration and concepting. Free tier available; Explorer at $29/mo.
Grain AI v3 — The meeting intelligence platform added a "meeting themes" feature that clusters recurring topics across all recorded meetings and surfaces patterns — which topics consistently run long, which questions get repeatedly deferred. Useful for managers who want to understand their team's meeting dynamics at aggregate rather than per-session. Pro at $19/user/mo.
Recraft v5 — The AI design tool for vector graphics and brand assets shipped v5 with improved logo generation and a "brand consistency" feature that generates multiple assets (icons, headers, social cards) as a coherent visual system from a single creative brief. Fills a gap between full design tools and simple icon generators. Free tier available; Pro at $20/mo.
Perplexity Finance — Perplexity launched a dedicated finance research mode on the Pro plan with live market data integration, earnings transcript search, and SEC filing analysis. Priced within the existing Pro plan ($20/mo), it is the most accessible entry point for AI-assisted equity research for individual investors and smaller funds.
Writer Enterprise v3 — Writer's enterprise content platform added house style enforcement that flags AI-generated content violating brand style guides and suggests corrections at the paragraph level. Relevant for large content operations producing AI-assisted content at volume where brand consistency is a QA concern. Enterprise pricing only; contact sales.
What to Watch in September 2027
September is shaping up to be another dense month. Adobe has signaled a Firefly 4 update for September with significantly improved image-to-image fidelity and a video generation feature that would put it in direct competition with Runway and Sora — if the quality matches the preview demos, it changes the video AI landscape again. GitHub Copilot Workspace is expected to exit beta and enter general availability, which puts enterprise agentic coding in front of every GitHub Enterprise organization almost by default. OpenAI has also teased a ChatGPT canvas update with multi-document editing and a research mode that sources against a connected knowledge base rather than the open web. The coding and video AI categories are heading into Q4 with the most competitive activity the market has seen — check back in September for the full roundup.
Quick Reference
| Tool | Category | Starting Price | Key August Feature | Best For |
|---|---|---|---|---|
| Runway Gen-4 | AI video | $15/mo | Consistent character, 90s clips | Scripted brand content |
| Cursor 2.0 | AI coding | Free / $20/mo | Background Agents, Project Memory | Full-stack dev teams |
| Figma AI | AI design | $15/user/mo (Pro) | Design Generation, Layers Cleanup | Product design teams |
| Claude 5 | AI model | $25/mo | Extended reasoning, 500K context | Research, legal, analysis |
| Notion AI 3.0 | Productivity | +$12/user/mo | Database Q&A, AI Automations | Ops teams in Notion |
| Elicit 3.0 | Research AI | Free / $20/mo | PRISMA review workflows | Academic researchers |
| Luma Dream Machine 2.0 | AI video | Free / $29/mo | 45s render, improved consistency | Rapid video concepting |
| Grain AI v3 | Meeting AI | $19/user/mo | Meeting theme clustering | Team managers |
| Recraft v5 | AI design | Free / $20/mo | Brand consistency system generation | Brand/marketing assets |
| Writer Enterprise v3 | Content AI | Enterprise | House style enforcement | Enterprise content ops |
| Perplexity Finance | Research AI | $20/mo (Pro) | Live market data, SEC filing search | Equity research |
Ready to discover the best AI tools every week? Subscribe to The DotProTools Digest — curated tools, practical workflows, delivered free every Tuesday.