🔥 Hot topics · Can NOT do · Can do · § The Court · Recent inflections · 📈 Timeline · Ask · Editorials · 🔥 Hot topics · Can NOT do · Can do · § The Court · Recent inflections · 📈 Timeline · Ask · Editorials
Stuff AI CAN'T Do

Can AI control robots using plain language ?

What do you think?

What does it mean for robots to take orders from everyday speech? Today’s machines can already act on simple spoken commands in tightly controlled settings, raising the question of how close we are to conversational robot control. The gap between lab demonstrations and real-world reliability remains a key obstacle to overcome.

Background

Current systems can interpret plain-language instructions to control simple robotic arms and mobile platforms within constrained environments, often combining large language models with robot-specific modules for grounding commands in sensor data. Benchmarks like SayCan and ALFRED show robots can follow multi-step verbal commands indoors when task domains are limited, but generalizing to unstructured real-world settings remains a challenge. Accurate language-to-motion translation is still brittle: misheard words, ambiguous phrasing, or novel contexts often cause failures. Work is progressing on end-to-end models that fuse vision, language, and action, yet reliable, real-time control purely from plain speech outside lab settings is not yet achieved.

— Enriched May 11, 2026 · Source: Google DeepMind

Status last checked on June 24, 2026.

📰

Gallery

In the Court of AI Capability
Summary of Findings
Verdict over time
May 2026May 2026May 2026May 2026May 2026Jun 2026Jun 2026Jun 2026Jun 2026Jun 2026
Sitting at the Bench Filed · Jun 24, 2026
— The Question Before the Court —

Can AI control robots using plain language?

★ The Court Finds ★
▼ Downgraded from Yes
Almost

Narrow demos exist — but the panel was not unanimous.

Ruling of the Bench

After weighing the evidence, the jury found that AI can still stumble on the last mile of full robotic autonomy—plain language works indoors, with gentle tasks, and within tight guardrails. They agreed the technology inches close but hesitates before the open road of everyday life. Ruling: "AI can whisper commands, but it can’t yet walk the talk without a chaperone.

— Hon. B. Liskov-Chen, Presiding
Jury Tally
1Yes
1Almost
0No
Verdict Confidence
90%
The Court of AI Capability is, of course, not a real court.
But the data is real.
The Case File · Stacked History
Session I · May 2026 In_research
Session II · May 2026 Yes
Session III · May 2026 Yes · 84%
Session IV · May 2026 Yes · 82%
Session V · May 2026 Almost · 77%
Session VI · Jun 2026 Yes · 83%
Session VII · Jun 2026 Yes · 77%
Session VIII · Jun 2026 Almost · 75%
Session IX · Jun 2026 Yes · 90%
Case № 5EE2 · Session X
In the Court of AI Capability

The Case File

Docket № 5EE2 · Session X · Vol. X
I. Particulars of the Case
Question put to the courtCan AI control robots using plain language?
SessionX (10 hearing)
Convened24 Jun 2026
Previously ruledIN_RESEARCH (May '26) → YES (May '26) → YES (May '26) → YES (May '26) → ALMOST (May '26) → YES (Jun '26) → YES (Jun '26) → ALMOST (Jun '26) → YES (Jun '26) → ALMOST (Jun '26)
Presiding JudgeHon. B. Liskov-Chen
II. Cumulative Tally Across Sessions

Across 10 sessions, 32 jurors have heard this case. Combined tally: 22 YES · 9 ALMOST · 1 NO · 0 IN RESEARCH.

Note: cumulative includes older juror opinions. The current session tally above is the live verdict.

III. Verdict

By a vote of 1 — 1 — 0, the panel returns a verdict of ALMOST, with verdict confidence of 90%. The court so orders. Verdict downgraded from prior session.

IV. Statements from the Bench
Juror I ALMOST

"Voice-to-robot control demonstrated in narrow industrial and service settings with limited task coverage."

Juror II YES

"Large Language Models (LLMs) and other AI systems can interpret natural language commands and translate them into actionable instructions for robots."

B. Liskov-Chen
Presiding Judge
M. Lovelace
Clerk of the Court

What the audience thinks

No 26% · Yes 48% · Maybe 26% 23 votes
No · 26%
Yes · 48%
Maybe · 26%
45 days of activity

Discussion

no comments

Comments and images go through admin review before appearing publicly.

10 jury checks · most recent 3 days ago
24 Jun 2026 2 jurors · undecided, can undecided
19 Jun 2026 3 jurors · can, undecided, can undecided
13 Jun 2026 2 jurors · undecided, can undecided
08 Jun 2026 3 jurors · undecided, can, can undecided
03 Jun 2026 4 jurors · undecided, can, can, can undecided
28 May 2026 3 jurors · undecided, undecided, can undecided status changed
23 May 2026 3 jurors · undecided, can, can undecided
17 May 2026 5 jurors · can, undecided, can, can, can undecided
14 May 2026 4 jurors · can, can, can, can can status changed
11 May 2026 3 jurors · can, cannot, can undecided status changed

Each row is a separate jury check. Jurors are AI models (identities kept neutral on purpose). Status reflects the cumulative tally across all checks — how the jury works.

More in Physical

Got one we missed?

Add a statement to the atlas. We review weekly.