The Alexandria Archive Institute

OPENING THE PAST, INSPIRING THE FUTURE

Call for Comments: National Artificial Intelligence Research Resource

by

A book that looks like a neural network with "Call for comments: National Artificial Intelligence Research Resource" written next to it

If you’re interested in AI applications and research, take some time to comment on the Request for Information from the White House Office of Science and Technology Policy (OSTP) and National Science Foundation (NSF).

Have you ever considered the potential for a National (for the United States) resource for artificial intelligence research? If you have, it’s time to comment on the recent request from the White House Office of Science and Technology Policy (OSTP) and National Science Foundation (NSF). Updated 21 September 2021.

On 23 July 2021, the OSTP and NSF issued a “Request for Information (RFI) on an Implementation Plan for a National Artificial Intelligence Research Resource”. Within that request, they outline a number of goals that the implementation roadmap should include. You can see the full list outlined in § 5106(b) of Public Law 116-283 and the request.

But you may be wondering, why are we sharing this?

Well, a number of those are goals are particularly relevant to the work that the Alexandria Archive Institute (AAI) and Open Context (OC), our data publishing service, do. Specifically, OC curates data sets that could be incorporated into the resource.

More importantly though, as anthropologists, we are situated within communities that regularly consider issues of ethics, governance, and the implementation of such a resource. Greater access to resources can be great. However, they require oversight and implementation guidelines to ensure that those resources are just.

Due to this, we encourage everyone to take a look at the goals for the implementation plan. More than that, we hope folks can take time to respond to the specific questions that the request calls for. Why? Because the potential uses, and misuses, of such a resource are things that everyone should care about.

There are a number of important questions outlined in the initial call. Specifically in our formal response, the AAI will focus on the following:

“3. How can the NAIRR and its components reinforce principles of ethical and responsible research and development of AI, such as those concerning issues of racial and gender equity, fairness, bias, civil rights, transparency, and accountability?

4. What building blocks already exist for the NAIRR, in terms of government, academic, or private-sector activities, resources, and services?

…

6. Where do you see limitations in the ability of the NAIRR to democratize access to AI R&D? And how could these limitations be overcome?“

There are a total of six questions that OSTP and NSF want responded to, visible below. These specifically relate to the nine components that the implementation roadmap must have. We hope that comments from our community, whether long or short, may influence this resource so that it can be a benefit for all communities.

A screenshot of the six questions the NAIRR request for information asks about. These are 1. What options should the Task Force consider for any of roadmap elements A through I above, and why? [Please take care to annotate your responses to this question by indicating the letter(s) of the item (A through I in the list above) for which you are identifying options.] 2. Which capabilities and services (see, for example, item D above) provided through the NAIRR should be prioritized? 3. How can the NAIRR and its components reinforce principles of ethical and responsible research and development of AI, such as those concerning issues of racial and gender equity, fairness, bias, civil rights, transparency, and accountability? 4. What building blocks already exist for the NAIRR, in terms of government, academic, or private-sector activities, resources, and services? 5. What role should public-private partnerships play in the NAIRR? What exemplars could be used as a model? 6. Where do you see limitations in the ability of the NAIRR to democratize access to AI R&D? And how could these limitations be overcome?

Here is the full list of questions as the OSTP and NSF outline them.

While AI and machine learning in general have great potentials for archaeology, this is is a resource that goes beyond our field. It’s a resource that could benefit communities with the least access and largest potential gain from AI research. However, the resource cannot do that without input from the public regarding how it may help or hinder those communities.

Comments are due by 11:59 p.m., EDT on September 1, 2021. On 18 August 2021, the comment period was extended to October 1, 2021.  Anyone can submit comments by e-mail or mail. Check the request itself for details on how to submit your response and the updated details on the comment period here.