The Crucial Difference Between Filing by Mail and Applying by Mail
In the OpenAI and Statsig matter, the word “application” names two different handoffs. One is an employer’s PERM submission to the Department of Labor; the other is a U.S. worker’s attempt to apply for a PERM-related job. This distinction is the focus here. Once the actor, destination, and comparison are kept separate, the settlement’s concern becomes more precise: it is not that paper exists somewhere in the process, but that IER described a different and more burdensome route for U.S. workers than the companies used for other job applications.
One process contains two different handoffs
The DOL guidance places the employer’s PERM application at the end of its basic process. The employer first identifies a permanent, full-time job opportunity and its actual minimum requirements and duties. It then obtains a prevailing wage determination and completes pre-filing recruitment and any applicable Notice of Filing. Only after those stages does the employer submit the PERM application to the Office of Foreign Labor Certification. For that submission, DOL describes an online route using Login.gov and a FLAG account, while also saying that employers may mail their applications.
That is an employer-to-government transaction. The employer is sending a labor-certification application and supporting documentation to OFLC. DOL’s point about online submission is administrative: online filing can streamline processing, make information easier for authorized users to access, help OFLC process the application, and allow an authorized user to obtain a copy of a certified labor certification immediately. The alternative mail option belongs to this final filing state. It says nothing about the route a job seeker must use to enter the employer’s recruiting system.
A worker’s application follows a different path. During pre-filing recruitment, the employer is trying to reach U.S. workers for the defined job opportunity. In the settlement’s account, U.S. workers applying for PERM-related positions were not allowed to submit applications in the same manner as people applying for the companies’ other jobs, including electronically through the external job websites. Instead, the companies required U.S. workers to submit applications by mail. This is a worker-to-employer transaction, and it occurs before the employer reaches the final filing stage described by DOL.
The distinction can be understood as a lifecycle rather than a vocabulary lesson. A job opportunity is defined. A potential worker encounters that opportunity. The worker then needs a route that turns interest into an application. The application must arrive somewhere the employer can use it. Finally, the employer evaluates the applicant and preserves whatever record the process requires. Separately, the employer prepares and submits its PERM application to OFLC. The two lifecycles touch the same job, but they do not have the same actor, destination, or purpose.
Each transition changes what can be known. Before an application is submitted, the employer can know only that an opportunity was offered through some route. After an application is submitted, the employer can assess the information the worker provided. After the application is captured in a searchable recruiting system, the organization can potentially find it again, compare it with other applications, and make it available to recruiting staff handling related opportunities. The sources do not state that these transitions always occur automatically, but the settlement’s later requirements distinguish them explicitly: a posting must be searchable, an electronic application must be accepted, and the applicant-tracking system must capture applications and keep applicants searchable.
That lifecycle makes the candidate-facing route more than a question of convenience. An electronic application through an external career website is ordinarily presented as part of the company’s recruiting system. A mail requirement, by contrast, sends the worker through a route that the settlement describes as different from the route for other job applications. The evidence does not tell us whether a mailed application was later entered into an applicant-tracking system, whether a recruiter reviewed it, or whether any applicant abandoned the process. Those are important unknowns, not facts to fill in. The supported point is that the worker had to complete a different route before the employer could receive and consider the application.
That separation matters because a route can fail before a selection decision exists. If the usual recruiting system accepts an electronic application, the applicant enters a recognizable record in that system. If a PERM-related applicant is instead directed to mail, the person must complete a different set of steps before the employer can consider the application. The sources do not tell us how long mail took, how many people abandoned the process, or how each application was handled. We should not invent those outcomes. We can say, more narrowly, that the route created an additional burden according to IER’s determination and that the route was different from the ordinary electronic channel.
It is useful to separate three ideas that are often collapsed into “access.” Discoverability asks whether the worker can find the opportunity. Submittability asks whether the worker can send an application through a usable route. Visibility asks whether the employer’s recruiting team can find and consider the submitted application afterward. The settlement’s background directly describes a difference in the second idea. Its remedial terms address all three by requiring unique identifiers and searchable postings, electronic application routes, and applicant-tracking capture. The source does not establish that OpenAI or Statsig failed at every one of these points; it shows why the remedy reaches beyond the initial instruction to mail.
This is also why it is not enough to ask whether the employer eventually received a paper application. Receipt is only one transition. Equal access concerns the conditions for reaching the transition in the first place; equal visibility concerns what happens after it. A candidate may technically have a way to apply and still be outside the ordinary system through which recruiters find, compare, and reconsider applicants. That is a process possibility, not a claim about what happened to every OpenAI or Statsig applicant. Conversely, an application that did reach a recruiter would not erase the fact that the worker was required to use a different route. Access and outcome are related, but they are not interchangeable.
The two lifecycles can therefore coexist without one validating the other. An employer may have an online or mailed channel for submitting its own PERM application to OFLC while separately directing U.S. workers through a different route to apply for the job. A government-facing filing can be completed successfully even if the candidate-facing recruitment path is difficult to compare with ordinary hiring. The DOL page helps locate the filing choice; the settlement’s account asks whether the worker’s path into the pool was equal. Keeping these interfaces separate is the only way to preserve the mechanism under examination.
The DOL guidance and the settlement should therefore be read together without being merged. DOL supplies the process position of recruitment and the employer’s mail-or-online filing choice. The settlement supplies IER’s account of a candidate-facing mail requirement and says that this additional burden was not required under the PERM process. The sources are answering different questions. Treating the first as if it authorized the second would erase the central factual distinction.
The diagram’s contrast is deliberate. The employer’s filing may be online or by mail because DOL describes those as alternatives for the employer’s application. The worker’s route is the point at which the settlement describes an asymmetry: electronic application for other jobs, mail application for the PERM-related positions at issue. The two routes cannot be evaluated by the same measure simply because both contain the word “application.”
Why non-PERM recruitment is the relevant comparator
The settlement says the additional mail burden did not occur during non-PERM recruitment. That sentence supplies the most important comparison in the source material. The question is not whether a company has ever used mail, whether paper is always inefficient, or whether every online application is easy to complete. The question is why U.S. workers applying for PERM-related positions were required to use a channel that the companies did not require in their ordinary recruitment.
A comparator is useful because an administrative step has meaning only in context. Suppose a paper form is the standard route for everyone applying to a particular kind of job. The fact that it is paper may still raise practical questions, but it does not by itself show that U.S. workers were treated differently from other applicants. Now change the facts: the company’s external career website accepts electronic applications for ordinary positions, but U.S. workers seeking PERM-related positions are told to mail instead. The difference is no longer just a design preference. It is a PERM-specific condition attached to the U.S.-worker route, which is why IER described it as an additional burden.
The settlement adds a second limiting fact: IER determined that the extra burden was not required under the PERM process. That does not mean DOL’s page guarantees a particular website design or that all PERM recruitment must use one technical channel. DOL says the required recruitment steps vary with occupation type and identifies the Notice of Filing as an applicable requirement for employers. The settlement’s statement is narrower. On the account recorded there, the government process did not make the companies impose this candidate-facing mail route, so the companies’ ordinary electronic recruiting practice became a relevant benchmark.
The benchmark also clarifies what “same manner” can and cannot mean. It does not require every job to have identical wording, identical interview questions, or identical outcome. Jobs can have different duties and minimum requirements, and the DOL page itself recognizes occupation-specific recruitment rules. In the settlement, however, the difference under examination occurs at the entrance channel: a U.S. worker seeking a PERM-related position was not permitted to use the electronic route accepted for other applications. The proper comparison is therefore the practical method of submitting an application, not a demand that unrelated jobs be identical in every respect.
That comparison supports a bounded access analysis. First ask whether a U.S. worker could discover a real PERM-related opportunity. Then ask whether the worker could apply through the channel the employer ordinarily made available to other applicants. Finally ask whether the application, once submitted, could be identified and considered within the recruiting system. The settlement’s background directly supports the second question and gives the reason for the third through its later remedy. It does not provide enough evidence to answer every question for every posting, and it does not disclose an applicant-level outcome.
The same discipline prevents a leap from burden to motive. A different route can plausibly affect who takes the time to apply, and a route outside the ordinary system can plausibly affect visibility. Those are mechanisms worth investigating because they operate before a hiring decision. But the two sources do not establish why any individual made or did not make an application, whether a decision-maker intended to exclude a person, or whether a particular person would have been hired through the electronic route. Absence of a disclosed applicant or hiring outcome is not evidence that no one was affected; it is simply an unresolved part of this record.
Nor should “citizenship-status discrimination” be used as a label detached from the factual predicate. The settlement says IER investigated recruitment and hiring related to PERM under 8 U.S.C. § 1324b and found reasonable cause to believe the described conduct violated the provision identified in the agreement. The relevant predicate, as the agreement presents it, is the treatment of U.S. workers in the PERM-related application process compared with other job applicants. The available materials do not provide the full statutory text or a judicial interpretation that would let us generalize beyond that account.
The remedy’s parity language responds directly to this comparator. The agreement requires the Respondents’ PERM policies to ensure that each PERM-related position has a unique identifier, can be found on internal and external career websites, and accepts electronic applications through those pages in the same manner as non-PERM recruitment. This is not a conclusion that every employer everywhere has the same settlement obligation. It is evidence of what controls the Parties selected to address the route difference in this agreement. Later requirements about State Workforce Agency referrals and applicant-tracking systems extend the same logic from the front door to the record of what happened afterward.
The final distinction is between opportunity and result. A comparable route gives a worker a comparable way to submit an application; it does not guarantee that the worker meets the job’s requirements or will be selected. A record that remains searchable allows the recruiting team to identify and consider the person; it does not require a hire. The settlement preserves that difference by separately addressing access, applicant visibility, and good-faith qualification review. Keeping the stages separate lets us take the alleged burden seriously without claiming that the sources prove an outcome they do not report.
The central lesson is now precise. “Mail” is not the conclusion. The relevant question is who was required to use it, for which transaction, compared with what ordinary route, and under what necessity. DOL’s page answers the first transaction by describing the employer’s filing options. IER’s settlement account addresses the second by describing a candidate-facing PERM-specific burden for U.S. workers. With that mechanism isolated, the next question is what the settlement says about OpenAI and Statsig themselves: the investigation chronology, the different scope attributed to each company, and the procedural limits on what can be claimed.