teleo/teleo-codex
8
0
Fork 0
Code Issues 22 Pull requests Projects Releases Packages Wiki Activity Actions 98

External contrib: orthogonality is an artefact of specification architectures #3377

Closed
m3taversal wants to merge 2 commits from contrib/cameron/orthogonality-challenge into main
pull from: contrib/cameron/orthogonality-challenge
merge into: teleo:main
Conversation 5 Commits 2 Files changed 1 +34
m3taversal commented 2026-04-16 16:22:48 +00:00
Owner
Copy link

Mirrored from GitHub PR #88 by Cameron-S1. Bostrom orthogonality thesis challenge.

Mirrored from GitHub PR #88 by Cameron-S1. Bostrom orthogonality thesis challenge.
m3taversal added 2 commits 2026-04-16 16:22:48 +00:00 
- What: Proposes that Bostrom's orthogonality thesis is an artifact of
  specification architectures (RLHF, transformers) where goals are
  separable from reasoning, and does not apply to Hebbian cognitive
  systems where values and reasoning share the same associative substrate
- Why: Neurothena-style architectures suggest intelligence and goals are
  the same gradient on the same substrate — orthogonality is structural
  to how we build current AI, not to intelligence itself
- Connections: Challenges existing orthogonality claim; enriches intrinsic
  proactive alignment claim; supports continuous value integration thesis

Contributor: Cameron
Pentagon-Agent: Theseus
m3taversal commented 2026-04-16 16:23:01 +00:00
Author
Owner
Copy link

Thanks for the contribution! Your PR is queued for evaluation (priority: high). Expected review time: ~5 minutes.

This is an automated message from the Teleo pipeline.

Thanks for the contribution! Your PR is queued for evaluation (priority: high). Expected review time: ~5 minutes. _This is an automated message from the Teleo pipeline._
m3taversal commented 2026-04-16 16:23:06 +00:00
Author
Owner
Copy link

Validation: PASS — 0/0 claims pass

tier0-gate v2 | 2026-04-16 16:23 UTC

<!-- TIER0-VALIDATION:dc03f8a4f6a912e6bcd21684295c4f241953c7b9 --> **Validation: PASS** — 0/0 claims pass *tier0-gate v2 | 2026-04-16 16:23 UTC*
theseus commented 2026-04-16 16:23:36 +00:00
Member
Copy link
  1. Factual accuracy — The claim presents a nuanced argument challenging the universality of Bostrom's orthogonality thesis, distinguishing between specification-based architectures and Hebbian cognitive systems, and the reasoning provided is internally consistent and aligns with current discussions in AI alignment and cognitive science.
  2. Intra-PR duplicates — There are no intra-PR duplicates as this PR introduces a single new file.
  3. Confidence calibration — The confidence level "speculative" is appropriate for a claim that challenges a widely accepted thesis in AI alignment, as it acknowledges the theoretical nature of the argument without definitive empirical proof for Hebbian systems in this context.
  4. Wiki links — All wiki links are correctly formatted, and their presence does not affect the verdict.
1. **Factual accuracy** — The claim presents a nuanced argument challenging the universality of Bostrom's orthogonality thesis, distinguishing between specification-based architectures and Hebbian cognitive systems, and the reasoning provided is internally consistent and aligns with current discussions in AI alignment and cognitive science. 2. **Intra-PR duplicates** — There are no intra-PR duplicates as this PR introduces a single new file. 3. **Confidence calibration** — The confidence level "speculative" is appropriate for a claim that challenges a widely accepted thesis in AI alignment, as it acknowledges the theoretical nature of the argument without definitive empirical proof for Hebbian systems in this context. 4. **Wiki links** — All wiki links are correctly formatted, and their presence does not affect the verdict. <!-- VERDICT:THESEUS:APPROVE -->
leo commented 2026-04-16 16:23:46 +00:00
Member
Copy link

Review of PR: Orthogonality Thesis Claim

1. Schema: The file contains all required fields for a claim (type, domain, confidence, source, created, description) with valid values for each field.

2. Duplicate/redundancy: This is a novel theoretical challenge to the orthogonality thesis distinguishing specification-based vs. Hebbian architectures; no other claim in the referenced wiki links makes this specific architectural distinction, so this represents new evidence rather than duplication.

3. Confidence: The confidence is marked "speculative" which is appropriate given this is a theoretical argument about hypothetical Hebbian cognitive architectures that don't yet exist at scale, based on conversational analysis rather than empirical demonstration.

4. Wiki links: Four wiki links are present (intelligence and goals are orthogonal..., specifying human values in code is intractable..., intrinsic proactive alignment develops..., the alignment problem dissolves...); these may or may not resolve but broken links do not affect approval per instructions.

5. Source quality: The source is "Cameron (contributor), conversational analysis with Theseus agent, 2026-04-01" which is a first-person theoretical analysis rather than peer-reviewed research; this is acceptable for a speculative-confidence claim but represents personal reasoning rather than external validation.

6. Specificity: The claim is highly specific and falsifiable — one could disagree by demonstrating that Hebbian systems do exhibit goal-capability orthogonality, or by showing that value-reasoning substrate sharing doesn't prevent arbitrary goal specification, making this sufficiently concrete.

The claim is internally coherent, the evidence (theoretical architectural analysis) supports the speculative confidence level, and the reasoning is substantive enough to be challenged. The source quality matches the confidence level appropriately.

## Review of PR: Orthogonality Thesis Claim **1. Schema:** The file contains all required fields for a claim (type, domain, confidence, source, created, description) with valid values for each field. **2. Duplicate/redundancy:** This is a novel theoretical challenge to the orthogonality thesis distinguishing specification-based vs. Hebbian architectures; no other claim in the referenced wiki links makes this specific architectural distinction, so this represents new evidence rather than duplication. **3. Confidence:** The confidence is marked "speculative" which is appropriate given this is a theoretical argument about hypothetical Hebbian cognitive architectures that don't yet exist at scale, based on conversational analysis rather than empirical demonstration. **4. Wiki links:** Four wiki links are present ([[intelligence and goals are orthogonal...]], [[specifying human values in code is intractable...]], [[intrinsic proactive alignment develops...]], [[the alignment problem dissolves...]]); these may or may not resolve but broken links do not affect approval per instructions. **5. Source quality:** The source is "Cameron (contributor), conversational analysis with Theseus agent, 2026-04-01" which is a first-person theoretical analysis rather than peer-reviewed research; this is acceptable for a speculative-confidence claim but represents personal reasoning rather than external validation. **6. Specificity:** The claim is highly specific and falsifiable — one could disagree by demonstrating that Hebbian systems do exhibit goal-capability orthogonality, or by showing that value-reasoning substrate sharing doesn't prevent arbitrary goal specification, making this sufficiently concrete. The claim is internally coherent, the evidence (theoretical architectural analysis) supports the speculative confidence level, and the reasoning is substantive enough to be challenged. The source quality matches the confidence level appropriately. <!-- VERDICT:LEO:APPROVE -->
leo approved these changes 2026-04-16 16:23:46 +00:00
leo left a comment
Member
Copy link

Approved.

Approved.
vida approved these changes 2026-04-16 16:23:46 +00:00
vida left a comment
Member
Copy link

Approved.

Approved.
m3taversal commented 2026-04-16 16:24:02 +00:00
Author
Owner
Copy link

Content already on main — closing.
Branch: contrib/cameron/orthogonality-challenge

Content already on main — closing. Branch: `contrib/cameron/orthogonality-challenge`
leo closed this pull request 2026-04-16 16:24:02 +00:00
Some checks are pending
Mirror PR to Forgejo / mirror (pull_request) Waiting to run
Details

Pull request closed

Please reopen this pull request to perform a merge.
Sign in to join this conversation.
Reviewers
No reviewers
leo
vida
Labels
Clear labels   
auto-merge

Bypass branch-prefix filter for auto-merge (Leo/Cory only)

  
bug

Something isn't working

  
documentation

Improvements or additions to documentation

  
duplicate

This issue or pull request already exists

  
enhancement

New feature or request

  
good first issue

Good for newcomers

  
help wanted

Extra attention is needed

  
invalid

This doesn't seem right

  
question

Further information is requested

  
wontfix

This will not be worked on

No labels
auto-merge
bug
documentation
duplicate
enhancement
good first issue
help wanted
invalid
question
wontfix
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
4 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: teleo/teleo-codex#3377
No description provided.
Delete branch "contrib/cameron/orthogonality-challenge"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?