Enhancing human intelligence with neurofeedback

Jan 19, 2025

Enhancing human intelligence with neurofeedback

Thomas Rialan, Nicole Dobreva, Anthony Kirilov

Read Project

Details

Summary

Build brain-computer interfaces that enhance focus and rationality, provide this preferentially to AI alignment researchers to bridge the gap between capabilities and alignment research progress.

Download PDF

View Presentation

Cite this work:

@misc {

title={

Enhancing human intelligence with neurofeedback

author={

Thomas Rialan, Nicole Dobreva, Anthony Kirilov

date={

1/19/25

organization={Apart Research},

note={Research submission to the research sprint hosted by Apart.},

howpublished={https://apartresearch.com}

}

Review

Reviewer Name

Constructive critique

Criteria 1:

Advancing interpretability (Possible questions to answer in this criteria: Does the project contribute to the field of mechanistic interpretability? Does it provide new insights into understanding or steering AI model behavior? How well does it move us towards reprogramming AI models? How original and innovative is the approach?)

2.5

Criteria 2:

Relevance for AI safety (Possible questions to answer in this criteria: How important is the contribution to advancing the field of AI safety? Do we expect the results to generalize beyond the specific case(s) presented in the submission? Does the approach introduce new safety mechanisms or enhance existing ones in innovative ways?))

2.5

Criteria 3:

Methodology & presentation (Possible questions to answer in this criteria: How well is the project executed from a technical standpoint?, Is the code well-structured, documented, and reproducible?, How effectively does it utilize Goodfire's SDK/API and other provided resources?, How clearly and effectively is the research presented in the paper?, Quality of visualizations and demos (if applicable), Clarity of methodology explanation and results interpretation))

2.5

Reviewer's Comments

Natalia Perez-Campanero

Interesting application of BCI! Curious about why you chose to focus on fNIRS in particular - the line of reasoning for this does not seem to be convincingly developed in your write-up. Given its effectiveness as a feedback modality is still a matter of debate due to its low spatial resolution and inconsistent signal quality amongst other concerns, it would be interesting to compare to alternative approaches (EEG/TMS/ECoG etc). It would also be valuable to think through the current competitive landscape in the space a bit more thoroughly, along with methodology both in terms of technical implementation and go-to-market strategy.

C1:

2.3

C2:

2.4

C3:

Finn Metz

The idea for the underlying is interesting, but also wasnt developed during the last weekend, and there is no demo/simulation. Would also somehow question if the "fooling oneself" is that much of an issue for researchers. I would imagine that this tech would be used for many other things, before being useful for detecting whether researchers are fooling themselves -> Please prove me wrong, I would love to have this not be the case. More straight forward path would probably to do BCI work in order to to come up with new mech interp approaches, but maybe this will become a byproduct further down the line. Obviously big dual use risk here.

C1:

1.6

C2:

1.2

C3:

1.1

Michaël Trazzi

I think trying to give a BCI product to AI Safety researcher is unlikely to make ai safety researchers much smarter than the rest, and don't see it scaling faster than other general BCIs. Lacks empirical exploration.

C1:

1.6

C2:

1.4

C3:

1.5

Recent Projects

View All

Mar 24, 2025

Attention Pattern Based Information Flow Visualization Tool

Understanding information flow in transformer-based language models is crucial for mechanistic interpretability. We introduce a visualization tool that extracts and represents attention patterns across model components, revealing how tokens influence each other during processing. Our tool automatically identifies and color-codes functional attention head types based on established taxonomies from recent research on indirect object identification (Wang et al., 2022), factual recall (Chughtai et al., 2024), and factual association retrieval (Geva et al., 2023). This interactive approach enables researchers to trace information propagation through transformer architectures, providing deeper insights into how these models implement reasoning and knowledge retrieval capabilities.

Mar 24, 2025

jaime project Title

bbb

Mar 25, 2025

Safe ai

The rapid adoption of AI in critical industries like healthcare and legal services has highlighted the urgent need for robust risk mitigation mechanisms. While domain-specific AI agents offer efficiency, they often lack transparency and accountability, raising concerns about safety, reliability, and compliance. The stakes are high, as AI failures in these sectors can lead to catastrophic outcomes, including loss of life, legal repercussions, and significant financial and reputational damage. Current solutions, such as regulatory frameworks and quality assurance protocols, provide only partial protection against the multifaceted risks associated with AI deployment. This situation underscores the necessity for an innovative approach that combines comprehensive risk assessment with financial safeguards to ensure the responsible and secure implementation of AI technologies across high-stakes industries.

Mar 24, 2025