PONE-D-20-16449

Mediating artificial intelligence developments through negative and positive incentives

PLOS ONE

Dear Dr. Han,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 31 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Alberto Antonioni, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

3. Please ensure that you refer to Figure 8 adn 9 in your text as, if accepted, production will need this reference to link the reader to the figure.

Additional Editor Comments (if provided):

The authors should address reviewers' constructive comments before considering the paper for publication in PLoS ONE. Overall, reviewers appreciated the work but give some useful suggestions to improve the presentation of the paper.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: N/A

Reviewer #3: N/A

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In the manuscript, the authors revealed the effects of rewards and punishments on AI developments (or other technology development races). They analytically and numerically solved the dynamics among some strategies (AS, AU, PS, and RS) by the means of Evolutionary Game Theory (EGT) and found that both rewards and punishments are effective for the developments.

I think the most contribution of this manuscript and Ref. 13 is that the authors invented the payoff matrix for the AI developments based on game theory. The payoff matrix is different from the one in social dilemma games, which EGT has frequently been applied. Although the original matrix was invented in Ref. 13 by the same authors, the authors included new strategies related to rewards and punishments in this manuscript. Thus, they successfully extended the study of this AI development games.

I recommend the publication of this manuscript in PLoS ONE after the following technical points are improved.

There are two points I find the authors can improve.

1. Organization of the manuscript. I think the authors can rearrange the structure of the manuscript to be better read.

Examples are below.

- The sixth and seventh paragraphs (lines 46-71) in the introduction can be combined with the fifth paragraph (lines 134-142) in the Materials and methods.

- When I read the model, especially Eq. 2., I forget what B and W are. Those variables are defined in the sixth paragraph (lines 46-57) in the Introduction, which is far from the model section. Perhaps, the Related Work section can be moved to the earlier part of the Introduction?

In short, if the authors could restructure the Introduction, Related Work, and Model, it would be beneficial to readers for better understanding.

2. Numerical calculations. It is better to give the detail procedures of the numerical calculation. I roughly understood how the analytical solutions were derived. However, I don't know how the numerical one is obtained. For example, in Figs. 2 and 4, the white lines are analytically obtained by calculating the fixation probabilities of strategies. A fixation probability represents the PROBABILITY that a single mutant replaces the existing population. However, the color maps in Figs. 2 and 4 obtained by the numerical calculations represents the FREQUENCY of the strategies. I don't know how the frequency is calculated. Did you numerically get the frequency by iterating Eq. 4 until the stationary distributions are obtained? If the authors give the detail procedure of the numerical calculations as a subsection, it would be great.

The followings are minor points.

- Could you add an intuitive explanation that why the risk-dominance thresholds are obtained only by two variables, p_r and s? I mathematically understand, if B/W is large enough, some variables can be ignored, but I don't know how it can be interpreted?

- Please add the clear definition of B. Lines 47 and 53 give the relation to that point but is not clearly defined. Does B imply the net benefit of the AI achievement?

- Line 315. "dynamics ." The space should be removed.

Reviewer #2: This is a very well written paper with a clear and concise model. My only suggestion would be to potentially enlarge the discussion. I believe the incentives discussed probably also apply to fields other than AI. Vaccine development or biotech in general, comes to mind.

I strongly recommend this paper is published.

Reviewer #3: This paper examines an evolutionary game theoretic model of a technological innovation race to study the effects of peer rewarding/punishment mechanisms. The technical part (mathematical modeling and analysis) seems sound and well presented, which should be worth publication in PLOS ONE.

Having said that, I think there are a few major issues in framing and presentation, which make the proposed model rather unjustified as a model of AI development. Details are explained below. The authors are strongly recommended to make a major revision in framing, presentation and justification of the model and its assumptions.

1. It was not clearly justified why the AI development is a particularly good interpretation of this model.

The model is a generic evolutionary game theoretic (EGT) model that represents competition among multiple players trying to reach a goal faster than the competitors while rewarding/penalizing the competitors according to their strategies. While this model setting itself is reasonably constructed and of some interest to the EGT community, it does not appear to describe the AI development scenario in particular. Actual AI technology development companies do not engage in a series of discrete pairwise game play events with their competitors (especially given that their development stages are asynchronous due to different development speeds), and they certainly would not have any incentives to impose the proposed peer rewarding/punishment (especially rewarding) on their opponents in the real industry ecosystem. These mismatches with reality make the model questionable as a model of AI development. There may be a better analogy for interpretation of this model (e.g., rapid vaccine development in which multiple competitors are also cooperating, to some extent, to achieve a global public health goal, for example).

2. The assumptions made about the speed of development do not seem to adequately capture the reality.

The key parameter of interest in this study is the speed of technology development, on which several assumptions were made: (1) The faster the speed is, the more unsafe the developed technology will be. (2) The speed of UNSAFE development is a fixed universal constant, and each developer has no control to adjust that speed, but is only able to choose either SAFE (slow) or UNSAFE (fast). (3) The competitor developer *can* change the opponent's speed by peer rewarding/punishment (even though it can't change its own speed in a similar way). None of these assumptions seems well justified, especially in the context of AI technology development. Unlike medical and pharmaceutical development that requires extensive clinical testing, technology development in computational domains can happen sometimes at a very rapid pace and result in a breakthrough, not necessarily in a risky technology. The speed of development is most likely not a GO-NOGO binary choice but should be more gradual strategic parameter that each developer can adjust by itself. Meanwhile, it would be rather difficult to influence the competitor's development speed from the outside, aside from filing lawsuits on, e.g., IP infringements (or perhaps by cyber-attacks).

Other miscellaneous points:

* Figure 2 was referred to earlier than Figure 1. The order of those figures should be reversed.

* It would be better to re-introduce definitions of symbols (S, B, W, p_r, etc.) in the Materials and Methods section.

* In page 8: "a cyclic pattern emerges (see Figure 3b)": Figure 3b does not show a cycle.

* In page 9: "see Figure 5, comparing panels c and f": Figure 5 does not have such panels.

* In page 9: "This observation confirms the observation": A redundant and unclear expression.

* Figure 1 does not have orange or blue areas.

* The direction of the blue line in Figure 1 does not look correct. Shouldn't its slope be negative, given that the rewarder reduces her speed?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Mediating artificial intelligence developments through negative and positive incentives

PONE-D-20-16449R1

Dear Dr. Han,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Alberto Antonioni, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The current version of the manuscript can be considered for publication on PLoS ONE as positively evaluated by the reviewers.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #3: The authors have addressed my comments (at least the limitations were discussed in the Discussions section). I recommend this manuscript to be published in PLOS ONE.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Genki Ichinose

Reviewer #3: No