home  /  the future of AI in legal  /  agentic systems & autonomy
strategy · ai for legal practice

Agentic systems and autonomy.

Responsibility for the output has not moved. The ability to inspect how it was produced has.

begin here

Where is your firm?

Start a conversation with the AI Adoption Concierge, already scoped to agentic systems & autonomy. Pick a starting point, or describe your situation directly.

AI Adoption Conciergeagentic systems & autonomy · orientation, not legal or ethics advice
Tell me what you are considering letting a system do end to end. I'll help you work out where a human has to stay in the loop for the work to be defensible.

An agentic system does not answer a question — it pursues a goal. It plans steps, runs searches, reads what it finds, drafts, checks its own work and calls other tools, looping until it decides it is done. For legal work the appeal is obvious, because a great deal of practice is exactly that kind of multi-step procedure. The difficulty is equally clear and it is not a technical one: professional responsibility for the output is unchanged, while the number of intermediate decisions a supervising lawyer would have to inspect to genuinely verify it has grown by an order of magnitude. The Law Society has named this directly as a widening gap between responsibility and auditability. It is not a transitional problem that better tools will quietly close.

mechanisms

The autonomy ladder.

Supervision is straightforward at the top and genuinely hard at the bottom.

Single response

One prompt, one answer. The lawyer sees exactly what was produced and reviews it. Conventional.

Guided sequence

The system runs defined steps in order. Predictable, and each step is inspectable.

Planned execution

The system chooses its own steps toward a stated goal. The plan itself becomes something to review.

Tool use

It calls search, documents and other systems. What it actually consulted is now a question with a non-obvious answer.

Self-correction

It reviews and revises its own output. Useful, and it means the reviewed version may hide the discarded reasoning.

Unattended completion

It finishes without a checkpoint. This is where responsibility and auditability come apart.

methodology

What the evidence shows — and what we examine.

How firms keep agentic work supervisable.

Mandatory checkpointsDefined points where a human must approve before the system continues. The single most effective control.
Full step loggingWhat it did, what it consulted, what it discarded — retained and readable, not just the final output.
Bounded scopeWhat the system may touch and what it may never do without a human. Written down before deployment.
A named supervisorOne person accountable per matter. Diffuse oversight is the same as none.
what's at stake

What autonomy changes.

Not the duty. Only the difficulty of discharging it.

what supervision can actually see whether review is genuine or nominal whether the process is reconstructable exposure when something goes wrong the size of the efficiency gain what supervisors need to be trained on

You remain responsible for output you cannot fully audit.

That is the gap, stated plainly. It is why the systems worth deploying are the ones that escalate decisions to a human and keep a readable record — not the ones that impress by needing nobody.

common questions

Agentic systems — practical questions.

Are firms actually using this?

Yes, and mostly in narrow, bounded applications rather than the general autonomy the term suggests. Several large firms have deployed agentic tooling internally for things like multi-step research, document processing pipelines and diligence review, where the steps are well-defined and the output lands in front of a lawyer. The gap between that and a system running a matter is very wide. Treat "we use agentic AI" as a statement about a workflow, not a description of autonomous practice, and ask what specifically it does unattended.

How is supervision supposed to work?

By designing checkpoints in rather than hoping review catches things at the end. The workable pattern is to identify the decisions that actually matter in a given workflow and require human approval at each, so the system does the volume and the lawyer makes the judgements. This is less efficient than full autonomy and it is the version that fits the professional obligations as they stand. A supervisor reviewing only a final output from a long autonomous chain is performing review in name, because the errors that matter are upstream of what they can see.

What should we log?

More than feels necessary, because the value only appears when something has gone wrong. At minimum: the instruction given, the steps the system chose, what sources it consulted, what it produced at each stage, and who approved what. The reason is that if the output is challenged — by a client, a regulator or a court — the firm needs to reconstruct how it was produced, and a log containing only the final answer cannot do that. Confirm the retention period is long enough to outlive the limitation period on the underlying matter.

Does this change our malpractice exposure?

It is a question to put to your carrier rather than one to answer from first principles, and it is worth putting now rather than after an incident. The standard is still what a reasonably competent lawyer would do, and that standard adapts to available technology in both directions — over time it may become unreasonable not to use certain tools, just as it is unreasonable to rely on them uncritically now. What is new is the evidentiary problem: demonstrating adequate supervision of a system whose intermediate steps were not recorded is very difficult, which is a practical argument for logging.

related

Related specialization areas & resources.

Deploying something agentic?

Describe what you want it to do unattended. The Institute will help you place the checkpoints.

AI adoption conciergeorientation · not legal or ethics advice
Tell me what you are considering letting a system do end to end. I'll help you work out where a human has to stay in the loop for the work to be defensible.