Intelligence= capability.
In pursuit of an abundant future, QWNTL Labs studies AI durability for both productivity and safety.
npx -y @qwntllabs/nenya setupThe intelligence is ready. Durability is the other half.
To achieve abundance, we must build an aligned and durable foundation, so that intelligence creates new productivity which grows in stride with human enlightenment.
Frontier models reason well enough for serious work, but they can’t be trusted in an enterprise setting, in public services, or even in managing your own life. What limits their use is the underdeveloped frontier of durability: the ability of agents to keep their goals, constraints and work valid as a project runs and changes.
We study safety as a precondition for productivity: an agent’s ability to be trusted is a product of it being safe to use, and dependable enough to be given work. We are working towards solving both by developing the model’s external “prefrontal cortex”.
In 1991, Shallice and Burgess studied three patients with injuries to their prefrontal cortex who scored well on intelligence tests. Sent on a simple errands task with its rules written on a card they carried, they regularly broke those rules. Their knowledge was intact. Executive function, the capacity to keep a goal and its constraints active over time, was not.
Long-running agents show the same gap, and we call the missing property durability.
In a 6,000-run comparison on three frontier models in our September 2026 paper, 97.6% of runs with the model’s external “prefrontal cortex” we developed finished with zero hard-constraint violations, against 18.4% for a leading agent framework running with memory and checkpointing.
Science is the most durable system we know, and we are building agents that work the way it does. Researchers leave; their methods, records and review outlast them, so each generation builds on checked work. Agents need the same property across sessions and model generations: capability that compounds, with its constraints and the means to verify it carried forward. Agents built that way could become how science itself gets done.
Read the thesisWhat we mean by durability
Long-horizon reliability
An agent can be trusted to take over projects with multiple stages of sub-tasks, reasoning, and decision making, over weeks and months.
Models as co-workers
An AI colleague learns how the project works, knows why decisions were made and acts when circumstances change.
Safe delegation
Less human intervention and hand-holding, so that human input is leveraged for creative or highly-complex strategic guidance, and standard decisions for project executions can be delegated to responsible models.
Systems of intelligence
We’re testing whether models and agents can work as one organisation, with discoveries that outlast the agents who made them. Similar to human societal learning over generational succession.
Sharp to the last decision
A model reasons as carefully late in the work as it does at the start, with the relevant evidence available for each decision, to ensure consistency of performance.
Knowing what is still true
New evidence changes the decisions that depend on it. The team stops using outdated guidance and acts on what applies now.
Research
Our research studies durability: how AI systems keep the rules they are given across long work, and how what they learn can outlast any one agent or model.
An experiment on a self-governing research organisation
We report an experiment on a self-governing research organisation of language-model agents, whose protocol admits a result only on hard evidence.
Measuring what durability adds to frontier models
We held the base model fixed and measured what durability alone changes over month-long professional work.
Can what an agent learns outlive it?
A study protocol for four worlds of a hundred agents each, to test whether knowledge and skill survive the agents that hold them.
Measuring when agents forget the rules they were given
Experiments on frontier models from Anthropic and Google show that an agent loses a rule at one moment: when it leaves the context window.
A technical summary of Nenya
The July results in five pages: the problems long-running agents meet, and what changes when their commitments are kept in force.
Why agents get pulled back from production
Agents are often withdrawn after they reach customers. We trace the cause to commitments that slip over the length of real work.
Why capability isn’t what limits long-running agents
Frontier models have capability to spare. What long, regulated work lacks is fidelity to the rules a model knows, held over time.
Beyond the Horizon: what durable AI could make possible.
The institution that invents its successors
A stronger successor model inherits decision provenance, including why past experiments were shelved and their identifiability limits, so it avoids rerunning studies that cannot distinguish the hypotheses.
A nervous system, rebuilt for a life
When an over-general learned rule fails on a new case, the system edits that rule locally, without catastrophic forgetting of the knowledge and methods that still hold.
Three founders build an energy industry
Procedures transfer across geothermal sites but site-specific conclusions do not; re-validating them on local evidence prevents negative transfer under distribution shift.
A river, over a century
A belief that held for fifty years fails under concept drift as droughts lengthen; the system revises it while keeping the measurement methods that detected the change.
An industry beyond Earth
An improvement found by a successor system is applied back to its predecessor, giving backward transfer across model generations.
1 hour · 1 day · 1 week · 1 quarter · 1 year · 1 decade · 1 century
Durable AI is the next great leap to models that can be trusted to learn, create, discover, and work alongside us.
Dawn of Durable IntelligenceSix problems between frontier models and durable AI
The six questions that decide whether agents can be trusted with long work, and the focus of our research.
See the six research areasAction state
Which evidence should govern the next decision, and what must change when that evidence changes?
Latent control
How can a model keep what it learns on the job through interruptions, restarts and new models?
Model continuity
What does a replacement model need from its predecessor to take over unfinished work?
Collective intelligence
Can a population develop abilities that survive its members and exceed what any member can do?
Self-improvement
Can a system use the methods it has learned to improve the AI that finds its next methods?
Economics
Which system finishes useful work for the least total effort, including the work people must repeat?
100 agents · 4 worlds · 30 simulated days
Capability that belongs to a population.
If an agent discovers a water source and later dies, can the group still use what that agent learned? The Valley is an experiment designed to find out: agents live in mortal bodies, sense only what is near them and leave records in objects. Can artificial intelligence properly form societal evolution under hostile survival conditions?
Read the protocol
We founded QWNTL Labs to build the most critical pillar for a safe and abundant future: scaled productivity in the service of humans.
Learn more about what influenced our dedication to durability and alignment.
Lindell CumesCo-founder and CEO 
Jacob SchoenbergCo-founder and Chief Growth Officer
Help build the key to abundance.
We want to hear from researchers and engineers who have pushed agents past a stable performance, whether that meant a months-long investigation or a team of agents that had to learn from each other. Tell us about the hardest problem you have hit and what you tried.