🔥 Hot topics · KAN IKKE · Kan · § The Court · Seneste omvendinger · 📈 Tidslinje · Spørg · Ledere · 🔥 Hot topics · KAN IKKE · Kan · § The Court · Seneste omvendinger · 📈 Tidslinje · Spørg · Ledere
Stuff AI CAN'T Do

Kan AI føre en flervejs samtale, der føles naturlig i ti minutter ?

Hvad mener du?

ChatGPT's udgivelse i november 2022 var det øjeblik, hvor den gennemsnitlige bruger kunne tale med en maskine og glemme i perioder, at de gjorde det.

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 senest tjekket July 3, 2026.

📰

Galleri

In the Court of AI Capability
Summary of Findings
Verdict over time
May 2026May 2026May 2026May 2026May 2026Jun 2026Jun 2026Jun 2026Jun 2026Jun 2026Jun 2026Jul 2026
Sitting at the Bench Filed · jul. 3, 2026
— The Question Before the Court —

Kan AI føre en flervejs samtale, der føles naturlig i ti minutter?

★ The Court Finds ★
▲ Upgraded from Næsten
Ja

Juryen fandt et klart bekræftende svar.

Ruling of the Bench

The jury found the AI convincingly capable of sustaining a natural, multi-turn conversation, praising its ability to retain context and respond in fluid, human-like exchanges over a significant duration. They concluded that while perfection remains elusive, the demonstrated fluency clears the threshold for this particular benchmark. Ruling: The chatbot has earned its dinner-party invite—for at least one course.

— Hon. B. Liskov-Chen, Presiding
Jury Tally
1Ja
0Næsten
0Nej
Verdict Confidence
98%
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 Næsten · 84%
Session IV · May 2026 Næsten · 79%
Session V · May 2026 Næsten · 79%
Session VI · Jun 2026 Næsten · 82%
Session VII · Jun 2026 Næsten · 79%
Session VIII · Jun 2026 Næsten · 79%
Session IX · Jun 2026 Ja · 94%
Session X · Jun 2026 Næsten · 86%
Session XI · Jun 2026 Næsten · 89%
Case № 74F7 · Session XII
In the Court of AI Capability

The Case File

Docket № 74F7 · Session XII · Vol. XII
I. Particulars of the Case
Question put to the courtKan AI føre en flervejs samtale, der føles naturlig i ti minutter?
SessionXII (12 hearing)
Convened3 jul. 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) → YES (Jul '26)
Presiding JudgeHon. B. Liskov-Chen
II. Cumulative Tally Across Sessions

Across 12 sessions, 31 jurors have heard this case. Combined tally: 15 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 — 0 — 0, the panel returns a verdict of JA, with verdict confidence of 98%. The court so orders. Verdict upgraded from prior session.

IV. Udtalelser fra dommerpanelet
Nævning I JA

"Leading LLMs pass Turing-style multi-turn conversational tests with sustained coherence."

Individuelle nævningers udtalelser vises på originalengelsk for at bevare bevismæssig præcision.

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

Hvad publikum mener

Nej 18% · Ja 70% · Måske 12% 106 votes
Nej · 18%
Ja · 70%
Måske · 12%
15 days of activity

Diskussion

no comments

Kommentarer og billeder gennemgår admin-godkendelse før de vises offentligt.

12 jury checks · seneste for 17 timer siden
03 Jul 2026 1 juror · kan kan
28 Jun 2026 2 jurors · kan, uafklaret uafklaret
22 Jun 2026 3 jurors · uafklaret, kan, uafklaret uafklaret
17 Jun 2026 2 jurors · kan, kan kan
11 Jun 2026 2 jurors · kan, uafklaret uafklaret
06 Jun 2026 2 jurors · kan, uafklaret uafklaret
01 Jun 2026 5 jurors · kan, kan, uafklaret, uafklaret, uafklaret uafklaret
26 May 2026 3 jurors · kan, uafklaret, uafklaret uafklaret
21 May 2026 2 jurors · kan, uafklaret uafklaret
15 May 2026 4 jurors · uafklaret, kan, kan, uafklaret uafklaret
12 May 2026 3 jurors · kan ikke, kan, kan ikke uafklaret
11 May 2026 2 jurors · kan ikke, kan uafklaret status ændret

Hver række er et separat jurytjek. Nævninger er AI-modeller (identiteter holdt neutrale med vilje). Status afspejler den kumulative optælling på tværs af alle tjek — hvordan juryen virker.

Flere i Relational

Har du en vi gik glip af?

Tilføj et udsagn til atlasset. Vi gennemgår ugentligt.