Research Grants

Academic Paper Grant

Whole Brain Emulation and Ab Initio AI Risks: an Integrated Picture

Research summary:

One route to machine intelligence is human Whole Brain Emulation (WBE), the careful scanning of human brains to create programs that emulate the structures of those brains at a fine-grained level and can mimic the input-output behavior of humans, including the ability to substitute for humans in essentially all cognitive tasks. This approach can be contrasted with attempts to produce Artificial General Intelligence (AGI) through more conventional software development, which might be inspired by study of the human brain in the sense that airplanes are inspired by birds but would not emulate it in fine detail. Opinions differ on whether WBE or non-WBE AGI will be developed first, and past analyses have mostly treated the two independently. However, there are several important connections between the two for purposes of risk analysis. WBE might be useful for reducing the risks posed by other sorts of AGI, or vice versa, and if so total risk might be reduced by influencing the relative timing of their development. Such efforts would themselves depend on development inputs shared by both, including computer hardware and neuroscience. This paper will analyze the existential risk implications of connections between WBE and non-WBE AGI.

Planned contents include:

  • An introduction of WBE and existential risks of AGI.
  • A discussion of the degree to which advancements in the power of computing hardware and advances in neuroscience support WBE and AGI respectively, and the the implications for the comparative development timelines of the two technologies; will knowledge of the brain provide insights to design AGI algorithms before brain emulation can be feasible, or will brute computational power and instrumentation continue to outpace AI theory, making brain emulation triumph as the comparatively 'low-understanding' solution?
  • An analysis of AGI risks and risk-reducing strategies given human brain emulations.
  • A discussion of the risks and potential of modification of brain emulations for enhanced cognitive faculties: effects on AI design capability, mental stability, and values.
  • An analysis of the use of limited AGI systems to produce brain emulation technology.
  • A discussion of uncertain parameters for which improved estimates offer high information value.
  • Implications for allocation of risk-reduction effort between affecting WBE or AGI development.

Prior related work:

The Future of Humanity Institute's Whole Brain Emulation Roadmap

If Uploads Come First, by Robin Hanson

At the post-Singularity Summit 2009 workshop, the vast majority of participants believed that developing WBE before non-WBE AGI would pose less existential risk than the reverse.

Target dates for:

Extended abstract (Posting an extended abstract on SIAI website; circulating to related academics for comment): 2 weeks after start date.[1]

 

Working paper (Posting a working paper on the SIAI website; circulating to related academics): 8 weeks after start date.

 

Conference submission: 14 weeks after start date.

 

Follow-up steps (Brainstorming, and drafting proposals for, any follow-up publications.  Should it be developed into a journal paper?): 15 weeks after start date.

 

[1] The "starting date" is the date (guaranteed to be within six months of the receipt of grant money) when we have skilled people to allocate to the project.  Extra donations increase our base of skilled people and thereby increase the number of projects we can get to; the lagged start date allows us to find new people, bring them here, and train them.


Total budget:  $5,900
  • Conference fees, air travel, motel: $1,400
  • Costs for researcher time: $4,500

How research costs are estimated:
  • Person-months for research and writing: 1.875 (This is obtained by taking our standard estimate[1] of 1.25 person-months per conference paper and multiplying by 1.5, since this paper requires gathering historical data.)
  • Dollars required to support one skilled full time researcher-month[2]: $2,400

[1] Our base estimate is 1.25 person-months per conference paper, and 3 per journal article, for an experienced full-time researcher. This estimate takes the planning fallacy, and the importance of an outside view in avoiding that fallacy, into account. While typical rates of article production by professors are extremely low, the distribution is strongly skewed towards research-oriented universities and departments, and informal surveys of researchers working on existential risks give data consistent with this estimate for full-time work required per paper. Visiting Fellows vary in their experience levels, so that mean productivity is expected to be lower, but a team mix can be selected to account for this.

[2] This billing rate reflects an estimate of financial outlays for SIAI to create the equivalent of one full-time skilled researcher-month, including stipend or hosting expenses, workspace, and administrative or management time, and other supporting expenses. Actual person-months may be greater or lower depending on the labor mix for a particular project, with shortfalls made up from general funds. This rate is not reflective of the money researchers could earn in the competitive labor market. Think of this as a matched donation. You donate the living expenses; our researchers donate the surplus value of their labor.

How this paper will help reduce existential risk:


Research benefits (What ideas will the paper explore?  How will that knowledge help with existential risk?)

  • This paper's topic is a core research question for evaluating the desirability and feasibility of altering the order in which WBE and non-WBE AGI systems are developed.
Influence benefits (What target audience will the paper impact, how?  How will that impact help with existential risk?)

  • The paper can enhance and be enhanced by the work invested in the Whole Brain Emulation Roadmap by the Future of Humanity Institute.
  • A series of analyses along the lines of this paper may eventually persuade funding agencies to invest in altering the order in which WBE and non-WBE AGI systems are developed.

Human capital benefits, or network benefits (Will writing this paper help new Visiting Fellows become familiar with key research domains?  Will it help create relationships with outside co-authors?  Will it give folks interested in existential risk entry into new communities where valuable contacts may be found?)

  • The paper offers an opportunity to collaborate with Future of Humanity Institute researchers and others who focused their analysis on brain emulations.
  • Visiting Fellows working on this paper will have a chance to look at high-level issues of strategy regarding AI risks.



Credit card transactions are securely processed through PayPal.

PayPal – Grant Donation

(200 characters max)