🔥 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 hold a multi-turn conversation that feels natural for ten minutes ?

What do you think?

What makes an AI chat feel truly human-like over an extended back-and-forth? While today’s systems can already converse, holding a natural ten-minute dialogue remains an elusive benchmark.

Background

ChatGPT’s release in November 2022 marked the first time many users could converse with an AI and, for stretches, forget they were interacting with a machine.

Current AI systems can engage in multi-turn conversations, but maintaining a natural flow for an extended period, such as ten minutes, remains a challenging task. While advancements in natural language processing have improved the ability of AI models to understand and respond to user input, they often struggle to keep track of context and adapt to subtle changes in conversation. As a result, conversations may start to feel forced or repetitive over time, lacking the nuance and depth of human interaction. Researchers continue to work on developing more sophisticated models that can capture the complexities of human conversation and sustain engaging interactions over longer periods.

— Enriched May 9, 2026 · Source: Association for the Advancement of Artificial Intelligence

Status last checked on June 28, 2026.

📰

Gallery

In the Court of AI Capability
Summary of Findings
Verdict over time
May 2026May 2026May 2026May 2026May 2026Jun 2026Jun 2026Jun 2026Jun 2026Jun 2026Jun 2026
Sitting at the Bench Filed · Jun 28, 2026
— The Question Before the Court —

Can AI hold a multi-turn conversation that feels natural for ten minutes?

★ The Court Finds ★
Reaffirmed
Almost

Narrow demos exist — but the panel was not unanimous.

Ruling of the Bench

The jury deliberated with near-unanimity on whether artificial intelligence can maintain a truly natural multi-turn conversation for ten minutes, with one juror holding out for a full pass. Though the lone dissenter argued that today’s models can feel eerily human, the rest agreed they occasionally stumble on tone or memory, leaving just a hair of imperfection in the verdict. Ruling: “Nine minutes of seamless chat, one minute of plausible doubt—verdict in the almost.”

— Hon. C. Babbage, Presiding
Jury Tally
1Yes
1Almost
0No
Verdict Confidence
89%
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 In_research
Session III · May 2026 Almost · 84%
Session IV · May 2026 Almost · 79%
Session V · May 2026 Almost · 79%
Session VI · Jun 2026 Almost · 82%
Session VII · Jun 2026 Almost · 79%
Session VIII · Jun 2026 Almost · 79%
Session IX · Jun 2026 Yes · 94%
Session X · Jun 2026 Almost · 86%
Case № 74F7 · Session XI
In the Court of AI Capability

The Case File

Docket № 74F7 · Session XI · Vol. XI
I. Particulars of the Case
Question put to the courtCan AI hold a multi-turn conversation that feels natural for ten minutes?
SessionXI (11 hearing)
Convened28 Jun 2026
Previously ruledIN_RESEARCH (May '26) → IN_RESEARCH (May '26) → ALMOST (May '26) → ALMOST (May '26) → ALMOST (May '26) → ALMOST (Jun '26) → ALMOST (Jun '26) → ALMOST (Jun '26) → YES (Jun '26) → ALMOST (Jun '26) → ALMOST (Jun '26)
Presiding JudgeHon. C. Babbage
II. Cumulative Tally Across Sessions

Across 11 sessions, 30 jurors have heard this case. Combined tally: 14 YES · 13 ALMOST · 3 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 89%. The court so orders.

IV. Statements from the Bench
Juror I YES

"Modern LLMs like GPT-4 or comparable systems sustain coherent, context-aware multi-turn conversation reliably."

Juror II ALMOST

"State-of-art chatbots can engage in lengthy conversations"

C. Babbage
Presiding Judge
M. Lovelace
Clerk of the Court

What the audience thinks

No 18% · Yes 70% · Maybe 12% 106 votes
No · 18%
Yes · 70%
Maybe · 12%
15 days of activity

Discussion

no comments

Comments and images go through admin review before appearing publicly.

11 jury checks · most recent 10 hours ago
28 Jun 2026 2 jurors · can, undecided undecided
22 Jun 2026 3 jurors · undecided, can, undecided undecided
17 Jun 2026 2 jurors · can, can can
11 Jun 2026 2 jurors · can, undecided undecided
06 Jun 2026 2 jurors · can, undecided undecided
01 Jun 2026 5 jurors · can, can, undecided, undecided, undecided undecided
26 May 2026 3 jurors · can, undecided, undecided undecided
21 May 2026 2 jurors · can, undecided undecided
15 May 2026 4 jurors · undecided, can, can, undecided undecided
12 May 2026 3 jurors · cannot, can, cannot undecided
11 May 2026 2 jurors · 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 Relational

Got one we missed?

Add a statement to the atlas. We review weekly.