🔥 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 generate a 3d model from a text prompt ?

What do you think?

Recent advances in generative modeling have pushed text-to-3D from early prototypes into real workflows, with systems now able to translate plain-language prompts into usable 3D assets. The panel notes that diffusion and NeRF-based techniques have matured to the point that once labor-intensive modeling tasks can be initiated with a single sentence.

Background

State-of-the-art text-to-3D systems now fuse diffusion priors with neural radiance fields to synthesize coherent meshes from prompts such as “a cyberpunk dragon on a neon-lit rooftop.” Public benchmarks from 2023–24 report FID scores around 30–40 when rendering novel views, indicating realism sufficient for rapid concept iteration rather than final production. Named models include DreamFusion (2023), which introduced Score Distillation Sampling to lift pre-trained 2D diffusion priors into 3D; followed by Magic3D (2023) that refines coarse NeRF outputs into high-resolution textured meshes in under an hour; and more recent approaches such as One-2-3-45 (2023) that go from a single image generated by a text prompt to a 3D model in about one minute. Limitations remain: fine geometric detail is often smoothed, prompting fails on abstract relations (“left of the red cube”), and outputs can collapse into degenerate geometries when longer prompts are used. These gaps are now the focus of techniques like multi-view diffusion guidance and per-prompt LoRA adapters.

SOURCE: Nature, 2024

Status last checked on September 26, 2026.

📰

Gallery

In the Court of AI Capability
Summary of Findings
Verdict over time
May 2026May 2026May 2026May 2026May 2026May 2026Jun 2026Jun 2026Jun 2026Jun 2026Jun 2026Jul 2026Jul 2026Jul 2026Jul 2026Jul 2026Jul 2026Aug 2026Aug 2026Aug 2026Aug 2026Aug 2026Sep 2026Sep 2026Sep 2026
Sitting at the Bench Filed · Sep 26, 2026
— The Question Before the Court —

Can AI generate a 3d model from a text prompt?

★ The Court Finds ★
▲ Upgraded from Almost
⚖
Yes

The jury found a clear answer in the affirmative.

Jury Tally
1Yes
0Almost
0No
Verdict Confidence
95%
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 · 86%
Session IV · May 2026 Yes · 86%
Session V · May 2026 Yes · 83%
Session VI · May 2026 Yes · 77%
Session VII · Jun 2026 Yes · 77%
Session VIII · Jun 2026 Yes · 86%
Session IX · Jun 2026 Yes · 82%
Session X · Jun 2026 Yes · 92%
Session XI · Jun 2026 Yes · 95%
Session XII · Jul 2026 Yes · 94%
Session XIII · Jul 2026 Yes · 95%
Session XIV · Jul 2026 Yes · 95%
Session XV · Jul 2026 Yes · 93%
Session XVI · Jul 2026 Yes · 90%
Session XVII · Jul 2026 Yes · 90%
Session XVIII · Aug 2026 Yes · 90%
Session XIX · Aug 2026 Yes · 93%
Session XX · Aug 2026 Yes · 92%
Session 21 · Aug 2026 Almost · 87%
Session 22 · Aug 2026 Yes · 95%
Session 23 · Sep 2026 Yes · 95%
Session 24 · Sep 2026 Almost · 85%
Case № 466B · Session 25
In the Court of AI Capability

The Case File

Docket № 466B · Session 25 · Vol. 25
I. Particulars of the Case
Question put to the courtCan AI generate a 3d model from a text prompt?
Session25 (25 hearing)
Convened26 Sep 2026
Previously ruledIN_RESEARCH (May '26) → YES (May '26) → YES (May '26) → YES (May '26) → YES (May '26) → YES (May '26) → YES (Jun '26) → YES (Jun '26) → YES (Jun '26) → YES (Jun '26) → YES (Jun '26) → YES (Jul '26) → YES (Jul '26) → YES (Jul '26) → YES (Jul '26) → YES (Jul '26) → YES (Jul '26) → YES (Aug '26) → YES (Aug '26) → YES (Aug '26) → ALMOST (Aug '26) → YES (Aug '26) → YES (Sep '26) → ALMOST (Sep '26) → YES (Sep '26)
II. Cumulative Tally Across Sessions

Across 25 sessions, 54 jurors have heard this case. Combined tally: 51 YES · 2 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 — 0 — 0, the panel returns a verdict of YES, with verdict confidence of 95%. The court so orders. Verdict upgraded from prior session.

IV. Statements from the Bench
Juror I YES

"Models like Luma AI's Genie and Meshy generate usable 3D assets from text prompts."

—
Presiding Judge
M. Lovelace
Clerk of the Court

What the audience thinks

No 3% · Yes 90% · Maybe 6% 62 votes
Yes · 90%
Trend needs votes from at least 2 different days.

Discussion

no comments

Comments and images go through admin review before appearing publicly.

⚖ 25 jury checks · most recent 20 hours ago
26 Sep 2026 1 juror · can can
15 Sep 2026 1 juror · undecided undecided
10 Sep 2026 2 jurors · can, can can
30 Aug 2026 1 juror · can can
19 Aug 2026 2 jurors · undecided, can undecided
14 Aug 2026 3 jurors · can, can, can can
08 Aug 2026 2 jurors · can, can can
03 Aug 2026 1 juror · can can
29 Jul 2026 1 juror · can can
23 Jul 2026 1 juror · can can
18 Jul 2026 2 jurors · can, can can
12 Jul 2026 1 juror · can can
07 Jul 2026 1 juror · can can
02 Jul 2026 2 jurors · can, can can
26 Jun 2026 1 juror · can can
21 Jun 2026 3 jurors · can, can, can can
15 Jun 2026 3 jurors · can, can, can can
10 Jun 2026 4 jurors · can, can, can, can can
04 Jun 2026 2 jurors · can, can can
30 May 2026 2 jurors · can, can can
25 May 2026 3 jurors · can, can, can can
19 May 2026 5 jurors · can, can, can, can, can can
15 May 2026 4 jurors · can, can, can, can can
12 May 2026 3 jurors · 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 Creative

Got one we missed?

Add a statement to the atlas. We review weekly.