EB-2 NIW National Interest Waiver guide for 2026.

An NIW is a petition that asks USCIS to waive the standard job offer and labor certification requirements because the applicant's work benefits the United States in a way that makes the labor certification process unnecessary. The waiver is discretionary — USCIS evaluates each case under the three-prong Dhanasar framework.

Step one: Prove EB-2 eligibility. You must show either that you hold an advanced degree (master's or higher, or a bachelor's plus five years of progressive experience) or exceptional ability in the sciences, arts, or business.

Step two: Prove the waiver is warranted. Under Dhanasar, you must show: (1) the proposed endeavor has substantial merit and national importance, (2) you are well positioned to advance it, and (3) on balance, waiving the labor certification benefits the United States.

Professionals with advanced degrees or exceptional ability whose work has a credible U.S. benefit. The NIW is available to self-petitioners and employer-sponsored applicants across virtually every field.

• Applicants who do not meet the EB-2 threshold — the NIW does not waive this.
• EB-1 applicants — different standards apply under Matter of Kazarian.
• Applicants considering priority date portability from an earlier I-140 — consult an attorney.

Prong one: Substantial merit and national importance. Substantial merit is broadly defined — USCIS has found substantial merit in fields as varied as biotechnology, education technology, business consulting, and the arts. The harder standard is national importance: the proposed endeavor must affect the field broadly, not merely benefit the petitioner or their employer. USCIS evaluates whether the endeavor has significance for the field as a whole, not whether the petitioner is the most accomplished researcher in their field.

Prong one evidence — tiered by strength:

• Tier 1 (direct adoption): Evidence that the applicant's work has been adopted, implemented, or cited by independent third parties. Examples: government agencies using the applicant's research methodology, federal grant award letters naming the applicant as PI or co-PI, licensed patents generating revenue or being used in commercial products, citation analysis showing independent citing institutions.
• Tier 2 (recognition by peers): Conference keynote invitations, journal editorial board memberships, peer review assignments from top-tier publications, national awards or fellowships (NSF CAREER, NIH New Innovator, DARPA Young Faculty), media coverage in national outlets that describes the specific work.
• Tier 3 (positional evidence): Affiliations with nationally prominent institutions, leadership roles in professional organisations, patents filed or granted (even if not yet commercialised), publications showing the arc of the applicant's contribution.

Note that Tier 3 evidence alone is rarely sufficient for Prong One. Patents that have never been cited, licensed, or produced revenue do not demonstrate national importance — they demonstrate invention, which is a separate factor. Successful NIW petitions typically combine evidence from at least two tiers.

Prong two: Well positioned to advance the endeavor. This prong evaluates whether the applicant has the skills, experience, and track record to execute the proposed endeavor. USCIS does not require the applicant to be the best in their field — the standard is whether they are well positioned, not whether they are uniquely positioned. Evidence categories include academic record showing relevant training, publication record with citation metrics demonstrating impact, patent assignments showing commercial potential, grant funding confirming that peer reviewers deemed the applicant's research worthy of funding, peer review invitations confirming standing in the field, expert letters from independent US-based researchers who describe specific contributions, and entrepreneurial evidence for startup founders such as incorporation documents, business plan, funding raised, and strategic partnerships.

Entrepreneur-specific evidence for Prong Two: If the applicant is self-petitioning based on a business or startup endeavor, Prong Two evidence should include: a detailed business plan with milestones, evidence of funding or investment (angel investor letters, venture capital term sheets, SBIR/STTR grant awards), customer letters of intent or contracts demonstrating market validation, and evidence the applicant has the relevant industry experience to execute the plan. USCIS evaluates whether the applicant personally is well positioned — the credibility of the business plan alone is not sufficient if the applicant's role and qualifications are unclear.

Prong three: On balance, waiver of job offer and labor certification benefits the United States. This prong asks: what does the US lose if this person must go through the PERM labor certification process? The answer connects timing, urgency, and field-specific need to the broader national interest. Evidence approaches include: a letter from a federal program officer confirming that a critical grant timeline depends on the applicant's continued work and would be disrupted by a 12-18 month PERM process, evidence that the field has a demonstrated shortage of qualified US workers supported by DOL data or industry reports, evidence that the endeavor addresses a national priority area identified in legislation or executive order (e.g., CHIPS and Science Act, AI Executive Order, NIH strategic plan), and an explanation of why the applicant's removal from the field would harm the national interest more than the PERM process would protect US workers.

The strongest Prong Three cases make a specific argument rather than a general one. General statements (”the field benefits from the applicant's work“) are not persuasive. Specific arguments (”the Department of Energy's Advanced Research Projects Agency — Energy (ARPA-E) has awarded the applicant's lab $2.8 million to develop grid-scale battery storage, and replacing the PI mid-grant would delay deployment timelines under the 2021 Infrastructure Investment and Jobs Act by at least two years“) carry weight because they tie the waiver request to a specific national priority with a documented timeline.

Organise evidence by Dhanasar prong, not by document type:

• Prong one: Citation analysis, federal funding, government contracts, national media, evidence of broad impact.
• Prong two: Credentials, work letters, publications, patents, awards, grants, expert letters, conference presentations, peer review history.
• Prong three: Letters explaining why labor certification would delay critical work, evidence of urgent national need, economic impact analysis.

Check the visa bulletin for priority date movement. If current, concurrent I-485/I-140 filing is available — significantly shortening the green card path. If not, file I-140 first. Premium processing (15-day) is available for NIW petitions.

The proposed endeavor description is the single most important document. It must be specific and written in plain English. Compare: ”I will research AI“ versus ”I will develop a federated learning framework for healthcare diagnostics, reducing data-sharing barriers while preserving patient privacy, as demonstrated by my published prototype in Nature Digital Medicine (2025).“ Each word of the second version generates a line of evidence: the framework itself (patent filed), the healthcare application (collaboration with a hospital system), the privacy-preserving feature (NSF grant supporting privacy research), the published prototype (publication record), and the journal name (citation prestige). A vague proposed endeavor denies the reviewer the ability to match evidence to claims.

Objective evidence outranks subjective praise. Citation metrics, grant award letters, and patent assignments are stronger than expert letters that call the applicant ”brilliant.“ This does not mean expert letters are unimportant — they are critical for Prong Two — but their weight depends on the specificity of the content. A letter that says ”Dr. X's algorithm reduced our diagnostic error rate by 12% in a trial of 3,000 patients, and we have integrated it into our clinical workflow“ is worth fifty letters that say ”Dr. X is an outstanding researcher.“

Evidence organisation strategy: Assemble evidence in three folders, one per Dhanasar prong, rather than grouping by document type. This makes the USCIS officer's job easier and reduces the chance that evidence for Prong One is overlooked because it was filed under a generic ”publications“ tab. Within each prong, lead with the strongest evidence — the independently verifiable fact (patent number, grant award number, citation count from Web of Science) rather than the subjective statement (letter of support). Use a table of contents or index that maps each document to the specific Dhanasar prong and element it supports. Many USCIS officers review NIW petitions by first reading the proposed endeavor description, then skipping to the evidence index, then spot-checking documents. If the index is clear, the review is faster and more likely to result in approval.

Amit is an Indian AI researcher with a PhD in computer science from the University of Illinois, currently on STEM OPT at a mid-sized AI startup. He has 4 first-author publications in top-tier conferences (NeurIPS, ICML), 85 citations from 14 US research hospitals and two federal agencies, 2 granted US patents on federated learning architectures, and a pending NSF grant as co-PI. He has no employer sponsor — he self-petitions for EB-2 NIW on his own behalf, funding his own legal representation and filing fees. His proposed endeavor: "Developing privacy-preserving federated learning frameworks that enable healthcare institutions to collaboratively train AI models without sharing patient data — addressing a critical barrier to medical AI adoption identified in the 2025 NIH Strategic Plan for Data Science."

EB-2 threshold: Amit's PhD in computer science satisfies the advanced degree requirement. He does not need to demonstrate exceptional ability because his advanced degree is sufficient, but his publication record also supports exceptional ability if challenged. His diploma, transcripts, and the university's accreditation documentation go in the EB-2 threshold section.

Prong 1 evidence — national importance:

• Tier 1: NSF grant award notification naming Amit as co-PI, which confirms federal peer review deemed his research approach valuable; citation analysis from Web of Science showing his federated learning papers have been cited by 14 US research hospitals, including three that name his framework as the basis for clinical AI pilot programs.
• Tier 2: Conference program committee invitations from NeurIPS 2025 and AAAI 2026; invited talk at the NIH Healthcare AI Symposium 2025; media coverage of his research in a national outlet covering the VA pilot program.
• Tier 3: Two US patent assignments held by Amit and assigned to the startup; his publication list demonstrating a consistent arc of research from theory (NeurIPS 2023) to applied framework (ICML 2025 with code release).

Prong 2 evidence — well positioned:

• PhD diploma and transcripts demonstrating advanced training in machine learning and federated systems
• Publication list with Google Scholar profile link showing h-index of 6 and the upward citation trajectory
• Patent assignments from the USPTO confirming Amit as sole inventor on two granted patents
• Three independent expert letters from US-based researchers: a professor at MIT who describes how Amit's framework improved her own federally funded research with a specific percentage improvement in model accuracy, a co-PI on the NSF grant who attests to Amit's role in designing the core algorithm, and a VA hospital researcher who adapted Amit's framework for the pilot program and can testify to its impact on patient outcomes

Prong 3 evidence — benefit of waiver:

• Letter from the NSF program officer confirming the grant timeline requires core technical work in year one (current calendar year) and that replacing Amit mid-project would delay deliverables beyond the grant period
• NIH Strategic Plan for Data Science (2025) showing that privacy-preserving collaborative learning is a designated priority area with specific deployment milestones
• Declaration from Amit explaining that the 12-18 month PERM labor certification process would require the startup to demonstrate a US worker shortage in the field of federated learning for healthcare — a sub-field so specialised that a PERM recruitment process is unlikely to produce qualified US applicants, but the process itself would delay the NIH-mandated deployment timeline beyond the grant period

USCIS approves the I-140 within 8 months with premium processing. Amit files I-485 concurrently when his priority date becomes current. The entire timeline from filing to green card is approximately 14 months — significantly shorter than the 2-3 years a PERM-based EB-2 would have required.

• Confusing career benefit with national importance. An endeavor that only helps the applicant or employer does not meet Dhanasar. National importance means the work affects the field broadly. Evidence that your work has been adopted by others (grantees, licensees, citing institutions) is stronger than evidence that your employer values it.
• Generic support letters. Letters should name specific projects and outcomes, not use general praise. A letter that says "Dr. X is brilliant" carries little weight. A letter that says "Dr. X's algorithm reduced our diagnostic error rate by 12% in a trial of 3,000 patients, and we have integrated it into our clinical workflow" demonstrates concrete impact.
• Weak EB-2 threshold. If the advanced-degree or exceptional-ability claim is unclear, USCIS issues an RFE. For exceptional ability claims, address at least three of the six regulatory criteria in 8 CFR 204.5(k)(3)(ii) with objective evidence, not assertions.
• Vague proposed endeavor. A plan too vague for Dhanasar evaluation will be denied. The endeavor should identify a specific problem, a concrete approach, and a measurable outcome. "I will work in AI" is not a proposed endeavor. "I will develop an open-source tool for detecting bias in clinical AI models, as prototyped in my 2025 NeurIPS paper, and distribute it through the NIH All of Us Research Program" is.
• Missing prong three. Some petitions cover prongs one and two but forget to explain why the waiver itself serves the national interest. Prong three asks: what does the US lose if this person must go through PERM labor certification? Connect the answer to urgency, a specific national initiative, or a demonstrated shortage of qualified US workers in the sub-field. A letter from a federal program officer confirming that a grant timeline depends on the applicant's continued work is strong Prong 3 evidence.