Attribution-Based Control in AI Systems
Andrew Trask's thesis argues that major AI risks stem from a lack of attribution-based control, which is caused by the overuse of addition, copying, and branching in gradient descent. The author proposes a new method synthesizing cryptography and distributed systems to enable bidirectional control between data sources and AI users. This approach aims to transform AI from a centralized intelligence tool into a communication platform that unlocks significantly more data and compute resources.
- ▪Many AI risks, including privacy issues and hallucinations, can be reduced to the problem of attribution-based control in AI systems.
- ▪The thesis identifies the overuse of addition, copying, and branching within gradient descent as the primary technical source of these control issues.
- ▪Proposed solutions involve synthesizing techniques from deep learning, cryptography, and distributed systems to provide formal guarantees for data usage.
- ▪Implementing attribution-based control would allow data sources to control which AI outputs they support and users to select specific data sources.
- ▪The author suggests this shift could unlock six or more orders of magnitude more data and compute while addressing geopolitical concerns about AI centralization.
Hacker News (AI / LLM) files mainly under ai. We currently carry 5,114 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Attribution-based-control |
| Canonical URL | https://attribution-based-control.ai/ |
| Publication time | Wed, 16 Sep 2026 07:08:16 +0000 |
| Retrieval time | 2026-09-16T07:28:41.353Z |
| Last seen | 2026-09-16T07:28:41.353Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | tNofF9BZuPRh · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Attribution-Based Control in AI Systems Andrew Trask University of Oxford · DPhil Thesis Pre-Print Abstract Many of AI’s risks in areas like privacy, value alignment, copyright, concentration of power, and hallucinations can be reduced to the problem of attribution-based control (ABC) in AI systems, which itself can be reduced to the overuse of addition, copying, and branching within gradient descent. This thesis synthesizes a handful of recently proposed techniques in deep learning, cryptography, and distributed systems, revealing a viable path to reduce addition, copying, and branching; provide ABC; and address many of AI’s primary risks while accelerating its benefits by unlocking 6+ orders of magnitude more data, compute, and associated AI capability Traditional open/closed-source AI…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Attribution-based-control.
Discussion
0 commentsMore from Attribution-based-control
