About

The consultancy born at the intersection of behavioral economics and human experience.

René Studio
AI EXPERIENCE STUDIO

The CX design platform we built from a decade of client work.

Explore René Studio
Mysterya
MYSTERY AUDITS, LIVE

Mystery audits we run, scored live for every location.

Explore Mysterya
NOW HIRING

Join a team reshaping how the world experiences brands.

View open roles →

COMPANY

GROW WITH US

CONNECT

Services

Comprehensive CX and management consulting for enterprise brands.

René Studio
AI EXPERIENCE STUDIO

Every engagement, mapped and scored in one AI workspace.

Explore René Studio
Mysterya
MYSTERY AUDITS, LIVE

Mystery shopping, scored and evidenced in real time.

Explore Mysterya
ALL SERVICES

Explore the full range of CX & management consulting services.

Browse all services →

CORE

SPECIALIST

Solutions

Structured solutions that turn CX ambition into measurable outcomes.

René Studio
AI EXPERIENCE STUDIO

Map, score and fix the journeys we redesign, with AI.

Explore René Studio
Mysterya
MYSTERY AUDITS, LIVE

Check the experience you designed is delivered.

Explore Mysterya
ALL SOLUTIONS

Explore every CX solution we offer.

Browse solutions →

STRATEGY & GOVERNANCE

DESIGN & DELIVERY

CULTURE & EXPERIENCE

Industries

A decade of CX transformation across the region's defining sectors.

René Studio
AI EXPERIENCE STUDIO

Sector-ready journeys, scored by AI in minutes.

Explore René Studio
Mysterya
MYSTERY AUDITS, LIVE

Mystery audits tuned to your sector, scored live.

Explore Mysterya
ALL INDUSTRIES

See how we work across every sector.

Browse industries →

BUILT ENVIRONMENT

FINANCE & TECH

PEOPLE & MOBILITY

Products

Proprietary tools, platforms, and AI that power CX transformation.

René Studio
AI EXPERIENCE STUDIO

Design, score and fix customer journeys with AI.

Explore René Studio
Mysterya
MYSTERY AUDITS, LIVE

Audit every location and act on what you find.

Explore Mysterya
REBELDECK A · 36 FORCES

The forces that shape how humans experience the world.

Explore REBEL Reveal →
ALL PRODUCTS

Explore the full Renascence product ecosystem.

Browse products →

PLATFORMS

AI & TECHNOLOGY

LEARNING & FRAMEWORKS

CX TOOLKIT

Opinion

Insights, research, and conversations at the frontier of CX.

René Studio
AI EXPERIENCE STUDIO

Turn what you read into a journey you can score.

Explore René Studio
Mysterya
MYSTERY AUDITS, LIVE

See what customers really experience, visit by visit.

Explore Mysterya
ReadExperience JournalArticles & research on CX, behavior, and transformation.Watch & listenExperience LoomOur video podcast on CX & behavior.CuratedCX NewsIndustry news that matters in CX, minus the noise.

Latest articles

Latest episodes

Latest news

Hub

Free tools, templates, and resources to advance your CX practice.

René Studio
AI EXPERIENCE STUDIO

Design, score and fix customer journeys with AI.

Explore René Studio
Mysterya
MYSTERY AUDITS, LIVE

Every visit, every location, every score. Live.

Explore Mysterya
THE MANIFESTOBurn the Deck.
Ten Virtues. Zero Excuses.Start reading →
THE HUB

Every free tool, template and resource in one place.

Visit the Hub →

AI TOOLS

FREE TOOLS

LEARNING

CULTURE

AI · 17 November 2026

AI Agent Memory: Why Knowing What to Forget Matters

Anthropic research suggests an AI agent's memory is judged more by what it discards than by how much it stores. In a Pokémon Red test, a newer Claude model progressed about four times as far as Sonnet 3.7 using a handful of strategy files.

Newsdesk
Curated briefing · 2 min read

What happened

Anthropic has published research suggesting that an AI agent's memory is judged less by how much it retains than by what it discards. In its study of how Claude carries knowledge across sessions, selective notes outperformed exhaustive archives by a wide margin, according to reporting by The Decoder.

The clearest illustration came from a Pokémon Red test. The newer Claude Fable 5 progressed roughly four times as far as the older Sonnet 3.7 by keeping just a handful of strategy files, rather than the more than 400 files that an archive-heavy approach would have held.

The research also describes a "Dreaming" process that revisits stored memories and retroactively corrects faulty ones. The available reporting does not go into detail on how that process is implemented in practice.

Why it matters

As AI agents move from single prompts to long, multi-step work, memory stops being a storage question and becomes a design decision. What an agent keeps, and what it quietly drops, shapes how consistently it behaves over time, how it handles changing instructions, and whether its outputs can be trusted by the people relying on them.

For organisations deploying agents in customer service, operations or internal workflows, the finding points towards governance rather than capacity. Teams may need to specify what an agent should retain, how stale or incorrect information gets corrected, and how these choices are reviewed, in the same way they would design retention and quality rules for any other system of record.

By the numbers

  • 4x the progress reported for Claude Fable 5 compared with Sonnet 3.7 in the Pokémon Red test
  • More than 400 files in the exhaustive memory approach, set against a handful of strategy files in the selective one

The Renascence take

Most coverage will read this as a model-performance story. The more useful reading is that forgetting is a feature of good memory, not a failure of it.

Teams building agents tend to treat memory like a warehouse, assuming more retained context means better decisions. This research suggests the opposite: relevance and curation do the heavy lifting, much as a good service agent remembers the one detail that matters rather than every past interaction. The "Dreaming" idea points to the same principle, because a memory that can be corrected is worth more than one that is merely large. A customer-obsessed operator should start by defining what its AI must never forget, what it should let go, and how errors in its memory will be found and fixed.

Sources

This briefing was written by our Newsdesk, synthesising reporting from the outlets below. Follow the links for the original coverage.

FAQ

Questions we get on this topic

According to reporting by The Decoder, Anthropic's study of how Claude carries knowledge across sessions found that selective notes outperformed exhaustive archives by a wide margin. The research suggests an agent's memory is judged more by what it discards than by how much it retains.

The newer Claude Fable 5 progressed roughly four times as far as the older Sonnet 3.7 by keeping just a handful of strategy files. An archive-heavy approach would have held more than 400 files.

What an agent keeps and drops shapes how consistently it behaves over time, how it handles changing instructions, and whether people can trust its outputs. Organisations may therefore need to set rules for what agents retain and how incorrect information is corrected.

Stay ahead of CX

Get the signal, not the noise.

The stories shaping customer experience — plus the Journal and Experience Loom — in your inbox.