INDEX

INDEX

RESEARCH

2026-05-09the model that can lie
2026-05-02agents need social memory
2026-04-25the invisible 98.8%
2026-04-04it can select, but it cannot create
2026-03-01dissecting hope: do self-modifying memories actually self-modify?
2025-12-15ai scientist

THINKING

2026-04-02lies of prediction
2026-01-21on breaking loops
2026-01-17the trajectory of a thread
2025-11-29connection and existence

ART

2026-03-25△w ∈ ℝⁿ

PRODUCT

2026-05-15token burn
2025-12ai arena

BLOG

multi-turn

 

2026-05-15token burn

invisible token consumption, visible in one menu bar line.

PRODUCT
2026-05-09the model that can lie

on separating truth from utterance.

RESEARCH
2026-05-02agents need social memory

Long-horizon agents need social state, not just log recall.

RESEARCH
2026-04-25the invisible 98.8%

I said one sentence. My words were 1.2% of the input.

RESEARCH
2026-04-04it can select, but it cannot create

AI recognizes what's funny, but can't create it.

RESEARCH
2026-04-02lies of prediction

on the lie that the predictable is the good.

THINKING
2026-03-25△w ∈ ℝⁿ

total museum ai art hackathon · p61 gallery berlin

ART
2026-03-01dissecting hope: do self-modifying memories actually self-modify?

do self-modifying memories actually self-modify?

RESEARCH
2026-01-21on breaking loops

the same thoughts travel the same path, and the path deepens.

THINKING
2026-01-17the trajectory of a thread

the trajectory of thread

THINKING
2025-12-15ai scientist

autonomous research agents

RESEARCH
2025-12ai arena

fluentdot.com

PRODUCT
2025-11-29connection and existence

we each live in a different universe.

THINKING

multi-turn inc.

south korea · est. 2025

© 2025 All Rights Reserved.

Multi-turn Inc. - Context is Intelligence