Document

Public Comments

ICR 202611-0693-001 · OMB 0693-0097 · Object 175043400.

Document Viewer [pdf]

Status: Original and derived artifacts are available for this document.

Download: pdf

Primary: pdfSource: application/pdf
Loading document viewer…

Document Metadata

Record metadata
application/pdf
Public Comments
Un-named
Writer
2026-10-20
complete

Extracted Text

Andrew G. Huff, Ph.D., M.S.
Risk Factor Research, LLC, Michigan, USA
[email protected]

October 20, 2026

National Institute of Standards and Technology (NIST), CHIPS Program Office, U.S.
Department of Commerce
OMB Control Number 0693-0097 (no regulations.gov docket)
Federal Register document 2026-17760; 91 FR 55843 (August 31, 2026)
Submitted by email to [email protected], subject line: Streamlined Supply Chain Information
Collection Request OMB Control Number 0693-0097; comments due October 30, 2026
Re: Agency Information Collection Activities; Comment Request; Streamlined Supply
Chain Information Collection Request; OMB Control Number 0693-0097; FR Doc. 202617760 (91 FR 55843, August 31, 2026)
I write in my individual capacity to comment on the proposed extension of the Streamlined
Supply Chain Information Collection Request, OMB Control Number 0693-0097, which
supports applications under Funding Opportunity 2023-NIST-CHIPS-SMME-01 for commercial
facilities producing semiconductor materials and manufacturing equipment. I am an
epidemiologist and security engineer. I previously served as a Senior Member of Technical Staff
at Sandia National Laboratories, where I did national security modeling of food systems, and I
have led or co-led federally funded programs across roughly ten agencies, so I know what it
costs an applicant to assemble the financial, technical, environmental, and workforce record this
collection asks for. My comments follow the four heads the Paperwork Reduction Act assigns to
this notice, with a short section, labeled as such, on handling of the supply-chain data collected.
Summary of recommendations
1. Reconcile the notice title with its text. The title describes a submission to OMB for review and
approval; the body states that the notice allows 60 days of public comment before the collection
is submitted to OMB. A reader who trusts the title may look for a comment link on reginfo.gov
that may not yet exist. One clarifying sentence in the 30-day notice, and consistent titling going
forward, would fix this.
2. State burden separately for the Concept Plan and the Full Application rather than as a single
blended figure of 42 hours per response across 210 respondents, and publish the assumptions
behind each.
3. Reconcile the $873,032 figure. The notice labels it the estimated total annual cost to the
public; the approved 2023 inventory record lists cost burden to the public as $0 and annual cost
to the Federal Government as $873,032. State which it is, and if it is a public cost, state the
hourly rate and labor categories used.
4. State which data elements changed, if any, between the 2023 approval and this extension, so
that the public can evaluate necessity and practical utility element by element rather than in the
abstract.

1

5. Minimize burden through technology by allowing applicants to reuse validated submissions
across stages, by accepting structured financial data in a published machine-readable format, and
by stating response-time targets so applicants can plan.
6. Beyond the PRA: bind provenance and validation status to every submitted value, publish the
handling rules for confidential business information submitted through the collection, and
confirm that the collection gathers nothing beyond what award evaluation requires.
The case for the collection as proposed
The strongest argument for extending this collection as it stands is that it works. The CHIPS
Incentives Program is authorized by Title XCIX of the Fiscal Year 2021 National Defense
Authorization Act, as amended by the CHIPS Act of 2022, and the program cannot evaluate an
application to construct, expand, or modernize a semiconductor materials or equipment facility
without the project, financial, environmental, and workforce information the abstract describes.
Respondents are for-profit businesses seeking federal funding, the obligation falls only on those
who choose to apply, and OMB approved the collection in November 2023 after public
comment. An extension of an active, approved collection serving a statutory program is exactly
the case the Paperwork Reduction Act treats as routine. My comments do not argue against the
collection. They argue that the notice should let the public check the agency's arithmetic and that
the agency should be explicit about how it protects what it collects.
(a) Necessity and practical utility
The collection is necessary. Section 4652 of Title 15 authorizes financial assistance for
semiconductor facilities, and no award can be made responsibly without the information the
abstract describes. Practical utility, however, is evaluated element by element, not at the level of
the whole instrument. The notice describes the categories collected (project descriptions and
timelines, justifications for incentives, applicant financial data, environmental and workforce
information), and I did not find in the notice a statement of whether any element was added,
removed, or revised since the 2023 approval. For an extension, that fact defines what is actually
being renewed. I recommend that the 30-day notice or the supporting statement include a short
change log: which elements are unchanged, which are new, and for each new element, the
award-evaluation decision it informs. An element that informs no decision has no practical utility
and should be dropped; one that does should be kept and defended on that basis.
(b) Accuracy of the burden estimate
The estimate is 210 respondents at 42 hours each, for 8,820 annual burden hours and an
estimated total annual cost to the public of $873,032. The hours arithmetic is internally
consistent; the problem is what the single blended figure hides. The approved inventory record
lists two instruments, a Concept Plan and a Full Application. A concept plan that gives a project
summary and a full application that includes financial data, environmental documentation, and
workforce information cannot plausibly take the same number of hours. The blended 42-hour
figure is either an average across very different instruments, which should be stated with its
weights, or an estimate for one instrument applied to both, which should be corrected. I
recommend separate response counts, hours per response, and cost per response for each
instrument, with the underlying assumptions published.

2

On the dollar figure, either the label or the methodology changed between the 2023 inventory
record and this notice, and the 30-day notice should say which. If $873,032 is a public cost,
dividing it by 8,820 burden hours implies about $99 per hour, a reasonable fully loaded rate for
the mix of finance, legal, engineering, and executive time a capital-project application consumes,
but the notice does not say that this is how the number was derived. I recommend stating the
labor categories, hourly rates, and any non-labor costs, such as third-party environmental or
financial review, that make up the total. From my experience assembling federal program
applications of comparable scope, 42 hours is at the low end for a full application that includes
financial data and environmental information. If the agency has actual completion data from
applicants since 2023, that data is the most probative evidence available and should replace the
original assumptions.
(c) Enhancing quality, utility, and clarity
Quality depends on respondents knowing what counts as a complete and valid answer. For each
data element, the instrument should state the acceptable form of the response, the validation the
agency will apply, and what happens when a value fails validation. This serves the applicant,
who can correct a submission before it is rejected, and the agency, whose reviewers then receive
data of known quality. In my work building surveillance and risk-modeling systems, a major
driver of downstream analytic cost was data whose validation status was unknown at the point of
entry; every value had to be re-checked because nobody could tell which values had already been
checked. Carrying a validation flag with each value costs little and removes that burden from
both sides.
Clarity would also improve if the notice gave applicants a stated response-time target for each
stage. An applicant who knows when to expect a decision can schedule the staff time the
collection consumes rather than holding it in reserve, which lowers the real burden without
changing the instrument.
(d) Minimizing burden, including through information technology
The collection is electronic, which is the right baseline. Three further steps would reduce burden
measurably. First, allow validated information submitted at the Concept Plan stage to carry
forward into the Full Application without re-entry, so that an applicant supplies each fact once.
Second, publish a machine-readable schema for the financial and project data so that applicants
can export from their own accounting and project-management systems rather than retyping, and
so that the agency receives structured data it can validate automatically. Third, treat interviews or
email, which the notice lists as alternative methods, only as supplements to the electronic record
and never as the system of record, since unstructured channels must later be reconciled with the
application at the agency's expense. Each step should be reported in the next supporting
statement with a before-and-after hours estimate so that the reduction is measured rather than
asserted.
Beyond the PRA: handling of supply-chain data
This section is outside the four statutory heads and is labeled as such. The collection gathers,
from a small number of firms, detailed information about domestic capacity to produce
polysilicon, photoresists, sputtering targets, deposition and etching tools, lithography and
inspection systems, and related inputs. Aggregated, that information describes the structure and
3

weak points of a national supply chain the CHIPS Act exists to strengthen. The inventory record
states that the collection contains personally identifiable information, and the notice warns
commenters not to submit confidential business information, which signals that the agency
understands the sensitivity involved.
I recommend three commitments, stated publicly. First, data minimization: collect only what
award evaluation requires, and treat each new internal recipient or secondary use of applicant
data as an expansion of the attack surface that must be justified by a stated purpose and a
measured benefit. Second, provenance and validation: bind to every stored value its source, the
date received, and its validation status, so that a reviewer, an auditor, or an incident responder
can tell a verified figure from an applicant assertion and can identify altered records. Third,
information-flow control: state where applicant data is stored, who can read it, how confidential
business information is segregated and protected under the Trade Secrets Act and Exemption 4
of the Freedom of Information Act, and how long it is retained after an award decision. NIST SP
800-161 Revision 1 on cybersecurity supply chain risk management, and the access-control and
audit families of SP 800-53 Revision 5, already supply the vocabulary for these commitments. I
made the same point about provenance and flow control in a separate comment on NIST's
National Vulnerability Database modernization, and it applies with equal force here: a record of
the nation's semiconductor supply base deserves the same handling discipline the agency's own
guidance describes.
Scope and confidence
I have not reviewed the supporting statement filed with OMB or the instruments themselves, and
I have not commented on the merits of the funding opportunity or any specific award. My
confidence that the blended burden figure obscures a material difference between the two
instruments is high, 80 to 90 out of 100, because the instruments differ in kind. My confidence
that the $873,032 label is inconsistent between the notice and the 2023 inventory record is high,
85 to 95; which document is correct I cannot say. My confidence that the 42-hour estimate
understates a full application is moderate, 60 to 75, and rests on experience with comparable
federal applications rather than on applicant data from this program, which the agency holds and
I do not. The data-handling recommendations rest on established NIST guidance and I hold them
with high confidence.
Disclosure: I have no financial interest in any vendor or contract involving this program, and I
was not compensated for this comment.
Thank you for the opportunity to comment.
Respectfully,
Andrew G. Huff, Ph.D., M.S.
Risk Factor Research, LLC
References
1. National Institute of Standards and Technology, Agency Information Collection Activities;
Submission to the Office of Management and Budget (OMB) for Review and Approval;
Comment Request; Streamlined Supply Chain Information Collection Request, 91 FR 55843,
4

August 31, 2026, https://www.federalregister.gov/documents/full_text/html/2026/08/31/202617760.html
2. Office of Information and Regulatory Affairs, ICR Reference No. 202310-0693-001, OMB
Control Number 0693-0097, CHIPS Streamlined Supply Chain Information Collection Request,
concluded November 29, 2023, https://www.reginfo.gov/public/do/PRAViewICR?
ref_nbr=202310-0693-001
3. Paperwork Reduction Act of 1995, 44 U.S.C. 3506(c)(2)(A),
https://www.law.cornell.edu/uscode/text/44/3506
4. Office of Management and Budget, 5 CFR 1320.8, Agency collection of information
responsibilities (Part 1320, Controlling Paperwork Burdens on the Public),
https://www.law.cornell.edu/cfr/text/5/1320.8
5. 15 U.S.C. 4652, Semiconductor incentives, https://www.law.cornell.edu/uscode/text/15/4652
6. Trade Secrets Act, 18 U.S.C. 1905, https://www.law.cornell.edu/uscode/text/18/1905
7. Freedom of Information Act, 5 U.S.C. 552(b)(4),
https://www.law.cornell.edu/uscode/text/5/552
8. National Institute of Standards and Technology, SP 800-161 Rev. 1, Cybersecurity Supply
Chain Risk Management Practices for Systems and Organizations, May 2022,
https://doi.org/10.6028/NIST.SP.800-161r1
9. National Institute of Standards and Technology, SP 800-53 Rev. 5, Security and Privacy
Controls for Information Systems and Organizations, September 2020,
https://doi.org/10.6028/NIST.SP.800-53r5

5