12 MAY 2026

8 New AI Tools: From deepfake takedowns to AI memory

Written by Wilco Verdoold

Our most recent batch of submissions highlights a different side of the AI market. Less customer-facing, more infrastructure, protection, and knowledge layers

This month’s selection includes tools for digital identity protection, legal redline analysis, AI-driven prospecting, integration infrastructure, business context for chat interfaces, persistent memory across assistants, agent security, and structured knowledge for long AI conversations.

For this selection, we picked 8 newly submitted tools that stood out for being practical, focused, or simply more original than the average AI clone. And yes, “stood out” is subjective. For transparent, objective overviews, check our monthly updated rankings.

Sidenty

Sidenty uses AI to detect and track leaked, manipulated, and unauthorised content online, including deepfakes, and links that monitoring to takedown and enforcement workflows.

Good for: creators, public figures, rights holders, platforms, and identity protection teams
Quick facts: Focused on deepfake detection, content protection, and takedown workflows; positioned around digital identity protection

Clausul

Clausul is a legal document comparison tool designed to help teams identify which contract changes actually matter instead of getting buried in formatting noise and endless redlines.

Good for: lawyers, legal ops teams, procurement teams, and contract reviewers
Quick facts: Focused on document comparison, redline workflows, and catching material contract changes more clearly

Relevate

Relevate is a documentation platform that generates API references from OpenAPI and includes an AI assistant that answers questions, writes content, and improves documentation based on user feedback.

Good for: API teams, technical writers, developer relations, and documentation maintainers
Quick facts: Built around API reference generation, AI documentation assistance, and content improvement workflows

Plumbed.io

Plumbed.io is an AI-powered integration platform designed to reduce launch time and maintenance overhead with prompt-driven setup and self-healing integrations.

Good for: technical teams, enterprise workflows, integration-heavy businesses, and data projects
Quick facts: Emphasises AI-powered integrations, lower engineering effort, self-healing behaviour, and faster deployment

CorpusIQ

CorpusIQ connects business tools to ChatGPT, Claude, and Perplexity so teams can query live business context with cited answers instead of jumping between disconnected apps.

Good for: operators, finance teams, analysts, founders, and AI-enabled business teams
Quick facts: Connects 25+ business tools; positioned as a private AI context layer with source-cited answers

MemoryLake

MemoryLake is a portable memory layer for AI interactions that preserves context across sessions, assistants, and tools instead of forcing users to start over each time.

Good for: agent builders, power users, AI workflow designers, and multi-tool AI users
Quick facts: Positioned as a “memory passport” across assistants; focused on persistent memory beyond single-session chat history

DKnownAI Guard

DKnownAI Guard is an AI agent security API focused on prompt injection detection, jailbreak defence, system prompt protection, and risk classification for agentic workflows.

Good for: AI product teams, security teams, agent builders, and developers
Quick facts: Built as a security layer for agentic AI; explicitly focuses on prompt injection detection and jailbreak defence

MindMarks

MindMarks helps users turn scattered AI conversations into structured knowledge by organising chats, generating navigation, and supporting exports across tools like ChatGPT, Gemini, and Claude.

Good for: researchers, writers, consultants, knowledge workers, and serious AI users
Quick facts: Positioned as a knowledge manager for AI conversations; focused on structure, reuse, and chat organisation