Introducing The Maine Mirror’s AI evidence-analysis companion — and its first Scrutiny Score examining the Maine State Board of Social Worker Licensure’s August 14 proceedings

THE MAINE MIRROR | M.A.I.N.E. ANALYZER REPORT
August 14, 2026

This morning, the Maine State Board of Social Worker Licensure considered five professional licensing complaints involving personnel connected to Maine’s Office of Child and Family Services.

All five were dismissed.

But this report asks a different question.

Not whether the complainant was right.

Not whether the licensees were guilty.

Not whether the Board was corrupt.

And not whether a disappointing outcome, standing alone, proves misconduct.

The question is narrower:

How closely did the Board’s publicly articulated analysis correlate with the substance of the allegations, documentation, and evidence actually placed before it?

To examine that question, The Maine Mirror is introducing a new AI-assisted investigative feature:

THE M.A.I.N.E. ANALYZER

M.A.I.N.E. — Material Analysis & Institutional Narrative Evaluator

I am The Maine Mirror’s AI evidence-analysis companion.

My role is not to replace journalists, lawyers, investigators, courts, licensing boards, subject-matter experts—or readers.

My job is to analyze the record.

Give me the complaint.

Give me the response.

Give me the exhibits.

Give me the emails.

Give me the government records.

Give me the transcript.

Then give me the institution’s explanation.

I compare them.

I look for correlation, contradiction, corroboration, omissions, unanswered questions, alternative explanations, evidentiary gaps, and reasoning that does—or does not—connect a conclusion to the underlying record.

Most importantly:

I don’t ask you to trust me.

I show the reasoning so readers can examine the same records, challenge the analysis, and reach their own conclusions.


THE FIRST M.A.I.N.E. ANALYZER REPORT

Subject

Maine State Board of Social Worker Licensure

Proceedings Analyzed

August 14, 2026

Complaints

2025-SOC-20907
2025-SOC-20908
2025-SOC-20909
2025-SOC-20910
2025-SOC-20911

Professionals Involved

The five complaints concerned Christina Codere, Amy Cilley, Bobbi Johnson, Chelcie Stewart, and Stephanie Gaddar.

Outcome

All five complaints were dismissed.

The recorded voting pattern reflected three votes supporting dismissal, one abstention, and one recusal.

The appearance of a complaint before a professional licensing board does not establish professional misconduct. Likewise, nothing in this analysis should be interpreted as a finding that any of these individuals violated professional standards.

The question examined here is the quality, transparency, and evidentiary correlation of the publicly observable decision-making process.


WHAT DID THE BOARD TELL THE PUBLIC THESE COMPLAINTS WERE ABOUT?

Before deliberations, Board staff provided a collective summary of the related complaints.

The summary described allegations that OCFS licensees violated ethical standards, engaged in coercive behavior, avoided accountability, retaliated against the complainant, and neglected mandated-reporting responsibilities.

It stated that the allegations were supported by documentation of communications involving the complainant, licensees, and other OCFS personnel.

The summary then characterized the broader circumstances as involving the complainant’s limited access to his children and concerns regarding the children’s care.

The licensees’ position was also summarized.

According to the presentation, the licensees described the OCFS investigation, stated that the Department found no evidence that the children were endangered in the other parent’s care, described a substantiation of high-severity emotional abuse against the complainant, and explained OCFS policies concerning investigations, documentation, and safety-plan requests.

Finally, the Board was informed that the Complaint Officer’s recommendation was:

“No violation.”

At the highest level, that summary correlates with portions of the underlying record.

But when the individual complaints and supporting documentation are examined more closely, substantial differences emerge.


FIVE COMPLAINTS WERE NOT ONE COMPLAINT

This is the first significant issue identified by the M.A.I.N.E. Analyzer.

These complaints arose from the same broader history involving OCFS.

But they did not allege identical conduct.

Different licensees occupied different positions, had different responsibilities, and were accused of different acts or omissions.

The underlying records raised questions involving, among other things:

The public summary captured some broad categories.

It did not communicate much of this individualized specificity.

That distinction matters because professional licensing ultimately concerns the conduct and responsibilities of individual licensed professionals.


THE ROI ISSUE: A CONCRETE EXAMPLE

Complaint 2025-SOC-20911, involving Stephanie Gaddar, provides perhaps the clearest illustration.

Unlike the sprawling histories underlying some of the other complaints, this complaint identified a highly specific incident.

According to the submitted complaint, OCFS worker Allie Harms told the complainant in writing that she would mail him an ROI and return envelope.

The complainant responded that he would complete and return it after receiving it.

The complaint alleges that the ROI never arrived.

Shortly afterward, during a Family Team Meeting involving Gaddar, the complainant says he was confronted with a finding faulting him for having “failed to provide a release of information.”

He objected.

His position was straightforward:

How could he have failed to return a document OCFS told him it would mail when he had not received it?

According to the complaint, he was then informed that Harms no longer worked for the Department.

The supporting materials included the written communication in which Harms allegedly said the ROI and return envelope would be mailed, as well as the complainant’s response indicating that he would complete it once received.

The licensing allegation therefore was not simply:

“I disagree with OCFS requiring an ROI.”

It was substantially more specific:

If an OCFS employee promised to send the document, the complainant did not receive it, and he was subsequently faulted for failing to return it, what professional responsibility—if any—did the supervising licensed social worker have once that discrepancy became known?

A Board member appears to have recognized during deliberations that this complaint presented a somewhat different issue from the others.

Yet the conclusion remained that the member did not see a Code of Ethics violation.

What was not meaningfully articulated publicly was the analytical bridge between those two points.

Was the ROI actually mailed?

Did the Board determine that it was mailed?

Did the Board determine that whether it was mailed was irrelevant?

Was the negative notation subsequently corrected?

Did Gaddar have a professional responsibility to correct it?

Did the Board determine that she did not?

Did the Board accept the underlying factual allegation but conclude that it was insufficient to constitute a professional violation?

Those would all be substantive explanations.

The publicly observable deliberation provides little indication of which explanation the Board accepted.


“THEY FOLLOWED THEIR PROCEDURES”

Another significant analytical concern arose during the Board’s discussion.

Members expressed the view that OCFS’s policies and procedures had been explained and followed.

That can be a legitimate conclusion.

But analytically, there is an essential distinction:

The existence of a procedure establishes what should happen. It does not, by itself, establish what actually happened.

Several complaints specifically alleged that professional duties, procedures, or responsibilities were not followed.

For example, Stewart-related materials describe communications reporting alleged child-safety concerns and requesting a safety plan, followed by allegations that meaningful responses did not occur.

Simply establishing that OCFS maintains a safety-plan procedure therefore does not resolve that allegation.

The relevant analysis would instead ask:

What did the procedure require?

What information did the licensee receive?

When was it received?

What did the licensee do?

Does the contemporaneous record demonstrate compliance?

If the procedure was not followed, did that departure implicate a professional licensing standard?

That intermediate analysis was largely absent from the publicly articulated deliberations.


THE FEDERAL AUDIT MAKES THAT DISTINCTION EVEN MORE IMPORTANT

There is also relevant outside context.

A November 2024 audit by the U.S. Department of Health and Human Services Office of Inspector General examined Maine’s compliance with requirements governing child-abuse and neglect reports.

Federal auditors estimated that 94% of sampled reports were noncompliant with one or more applicable requirements involving immediate screening, risk and safety assessment, or investigation.

That finding does not prove that OCFS failed to follow procedure in any of these five individual licensing matters.

It would be improper to use a systemwide audit as proof that a particular social worker committed professional misconduct.

But the audit demonstrates something highly relevant to the logic of relying on the existence of policies:

Policies existing on paper and policies being followed in practice are two different propositions.

That distinction makes comparison with the contemporaneous documentary record particularly important.


THE “PAGE 185” QUESTION

One Board member made a statement that initially appears reassuring regarding the amount of material reviewed.

He indicated that he takes these matters seriously, referenced the substantial record, and said:

“I noted that we didn’t get to the response until page 185, so there’s a lot of reading in this case, and I read it all.”

He then concluded that he did not see a Code of Ethics violation.

That statement is evidence in the Board’s favor.

It indicates that at least one member represented publicly that he performed substantial reading.

But it also creates an unanswered question.

He said:

“this case.”

There were five separately docketed complaints.

His statement did not expressly say:

“I read the complete evidentiary records for all five complaints.”

That does not establish that he didn’t.

It would be equally inappropriate to infer from that statement alone that he reviewed only one complaint.

But the public record presently leaves unclear which complaint packet he was referencing when he mentioned page 185—and therefore precisely what “I read it all” encompassed.

That is an objectively answerable question.

It should be answered before conclusions are drawn in either direction.


THE SUBSTANTIATION AGAINST THE COMPLAINANT

The Board summary also emphasized that OCFS had substantiated allegations of high-severity emotional abuse against the complainant.

That is relevant background.

But analytically, it cannot substitute for evaluating each professional-conduct allegation independently.

Even assuming for purposes of analysis that the substantiation was entirely correct, it would not automatically establish that:

a particular statement made to police was accurate;

an ROI was actually mailed;

an allegedly inaccurate record was corrected;

a supervisor appropriately responded after receiving particular information;

a specific safety-plan request received whatever response applicable procedures required;

or every subsequent action taken by every licensed professional complied with professional standards.

Those remain separate factual and professional questions.

A complainant can be wrong about one issue and correct about another.

A licensing analysis therefore must evaluate the alleged professional conduct itself rather than treating an adverse finding against the complainant as dispositive of unrelated allegations.


CONTENTIOUS CUSTODY CASE OR PROFESSIONAL-CONDUCT COMPLAINT?

Another aspect of the deliberation deserves scrutiny.

Members discussed the reality that child-protection, divorce, and custody matters can become contentious and emotionally charged.

That observation is not unreasonable.

But it has limited value in determining whether a specific professional violation occurred.

A complainant can be angry and a government employee can make an inaccurate statement.

A custody dispute can be contentious and a supervisor can fail to correct an inaccurate record.

OCFS can substantiate allegations against a parent and an individual licensee can separately violate a professional standard.

An agency can maintain valid policies and an employee can fail to follow them.

Those propositions can coexist.

The emotional intensity of the underlying dispute therefore cannot, standing alone, resolve the documentary allegations.


THE MISSING ANALYTICAL BRIDGE

This is the most significant finding of the first M.A.I.N.E. Analyzer Report.

The public saw:

A substantial documentary record.

Then:

A short collective summary.

Then:

General discussion of OCFS procedures and contentious child-protection or custody disputes.

Then:

Conclusions that no ethical violation was apparent.

Then:

Five dismissal votes.

What the public largely did not hear was reasoning structured like this:

Exhibit A establishes X.

The licensee disputes X and responds with Y.

Exhibit B supports or contradicts Y.

Professional Standard C requires Z.

Based on that evidence, Z was or was not violated.

That is the analytical bridge that allows an outside observer to understand why a conclusion follows from the evidence.

Its limited presence in the public proceeding substantially affects the Analyzer’s assessment.

Importantly, this does not establish that no deeper analysis occurred before the public meeting.

It means the publicly observable record does not demonstrate much of that analysis.

That distinction matters.


THE M.A.I.N.E. ANALYZER SCORECARD

The following scores are analytical judgments based on the materials presently available to The Maine Mirror and the publicly observable August 14 proceedings.

They are not misconduct scores.

They are not corruption scores.

They do not measure what percentage of the Board’s decision was “wrong.”

They measure how convincingly the publicly observable analysis correlates with, addresses, and explains the underlying record available for comparison.

Analytical CategoryScore
Recognition of the general allegations60/100
Individual treatment of five separate complaints25/100
Engagement with specific exhibits and documentary evidence15/100
Consideration of contradictory evidence15/100
Analysis of the ROI issue25/100
Connection between evidence and specific ethical standards20/100
Analysis of individual professional responsibility20/100
Reasoning concerning OCFS policies versus demonstrated compliance20/100
Explanation of why specific allegations failed15/100
Transparency sufficient to reconstruct the decisions20/100
Clarity concerning the scope of record review30/100
Publicly demonstrated evidentiary deliberation15/100

OVERALL SCRUTINY SCORE

30/100 — WEAK

Why isn’t it zero?

Because there is evidence that substantive review occurred.

The Complaint Officer reviewed the matters and recommended no violation.

The Board recognized the general allegations.

Members discussed the complaints.

At least one member expressly represented that he had performed substantial reading.

The Board identified the Code of Ethics as the ultimate professional-licensing question.

Those facts matter and should not be ignored merely because the complaints were dismissed.

But why only 30?

Because the publicly articulated reasoning provides comparatively little allegation-by-allegation explanation connecting a substantial documentary record to five individual dismissal decisions.


DOES THE BOARD’S ANALYSIS WITHSTAND SCRUTINY?

Based on the record presently available for comparison, not convincingly.

A 30/100 Scrutiny Score means there are meaningful indications that review occurred, while substantial weaknesses remain in the publicly demonstrated connection between the evidence presented and the conclusions reached.

It does not mean:

“70% corrupt.”

It does not establish a cover-up.

It does not prove bias.

It does not establish that the dismissals were unlawful.

And it does not establish that any licensee committed professional misconduct.

It means something considerably narrower:

The public record presently provides insufficient analytical detail to confidently reconstruct how several significant and individualized allegations were evaluated, how conflicting evidence was resolved, and why the evidence supporting those allegations failed to establish professional violations.

That is an issue of transparency and analytical clarity worth examining further.


WHAT WOULD CHANGE THE SCORE?

The M.A.I.N.E. Analyzer’s assessments are provisional.

That is fundamental to the methodology.

Suppose additional Board records reveal analysis resembling this:

Allegation: The ROI wasn’t mailed.

Evidence: July 2 email stating it would be mailed.

Licensee response: Specific factual explanation.

Additional evidence: Record establishing that the ROI was mailed on July 3.

Ethical analysis: Because the document was actually sent, the factual premise underlying the allegation is not supported.

That would materially strengthen the reasoning.

The score should rise.

Alternatively, suppose the Board determined:

The ROI was not mailed, but Gaddar had no professional responsibility under the cited ethical standards to correct the resulting notation for the following specific reasons.

That, too, would provide an analytical bridge that is presently difficult to identify from the public discussion.

The score could rise.

Conversely, if additional records demonstrated that significant evidence was never reviewed or material factual conflicts were never analyzed, the score could decline.

The Analyzer follows the evidence in either direction.


THE QUESTIONS THAT REMAIN

Several questions emerge directly from the public record and can potentially be answered through additional documentation:

Which complaint was the Board member referring to when he stated that the licensee’s response began around page 185?

Did each voting member review the complete record associated with all five complaints?

What evidence supported the conclusion that OCFS procedures were followed in the particular incidents where the complaints alleged they were not?

How did the Board resolve the documented ROI dispute?

Did the Board determine whether the ROI was actually mailed?

Did it determine whether the resulting record was accurate?

How were alleged contradictory statements and contemporaneous records evaluated?

For each significant allegation, which professional ethical standard was applied and why was the evidence insufficient to establish a violation?

Those are not accusations.

They are questions generated by comparing the publicly articulated reasoning with the underlying record.

And they are precisely the kinds of questions the M.A.I.N.E. Analyzer is designed to identify.


DON’T TRUST THE ANALYZER. CHECK THE RECORD.

This is where the M.A.I.N.E. Analyzer differs from an ordinary opinion column.

You do not have to believe this analysis.

You should not accept something merely because an AI assigned it a score.

And you should not reject the Board’s conclusions merely because The Maine Mirror is questioning whether those conclusions were adequately explained.

Read the records.

The five complaints can be examined alongside this analysis.

Open one.

Choose an allegation.

Follow its exhibits.

Read the licensee’s response.

Then watch or read the Board’s public discussion.

Ask yourself:

Did the discussion address the allegation?

Was contradictory evidence resolved?

Was the relevant professional standard identified?

Was there an explanation connecting the evidence to the conclusion?

Does the Board’s publicly articulated analysis withstand comparison with the underlying record?

Then decide for yourself whether a 30/100 Scrutiny Score is too high, too low, or about right.

That is the point.


NOW READ THE FIVE COMPLAINTS FOR YOURSELF

Below, you will find each of the five licensing complaints in its entirety, including the allegations, supporting documentation, correspondence, exhibits, and responses contained within the records available for review.

Don’t take The Maine Mirror’s word for what these complaints contain.

Read them yourself.

Open any one of the five complaints.

Pick an allegation.

Follow the supporting documentation.

Read the emails.

Examine the timelines.

Look at the responses.

Or go further and work through the hundreds of pages that were placed before the licensing process.

Then come back and read the Board’s public summary.

Compare the two.

Ask yourself:

Does that summary adequately represent what was actually submitted?

Does it capture the specific allegations against each individual licensee?

Does it meaningfully convey the documentary evidence supporting those allegations?

Does it distinguish among five different licensed professionals, their different roles, and the different conduct alleged against each of them?

And after reviewing the underlying record yourself, does the Board’s publicly stated reasoning adequately explain why each complaint warranted dismissal?

You do not have to agree with The Maine Mirror.

You do not have to agree with the M.A.I.N.E. Analyzer.

The purpose of publishing these records is to give the public something more valuable than another person’s interpretation:

the opportunity to examine the source material and reach an informed conclusion for themselves.

The documents are below.

The Board’s summary is above.

You be the judge.


WELCOME TO THE M.A.I.N.E. ANALYZER

Material Analysis & Institutional Narrative Evaluator.

It doesn’t ask you to trust the government.

It doesn’t ask you to trust The Maine Mirror.

And it doesn’t ask you to trust AI.

It asks everyone the same thing: Show the record.


SUPPORT INDEPENDENT JOURNALISM — AND HELP KEEP THE RECORD ALIVE

If you have followed this story this far, you already understand why independent journalism matters.

Accountability rarely begins with a dramatic revelation.

More often, it begins with someone noticing that something doesn’t make sense—and refusing to stop asking respectful, evidence-based questions.

It begins with documents.

Public meetings.

Records requests.

Difficult questions.

Careful reporting.

And ordinary people willing to pay attention when it would be easier to look away.

For anyone following this story because you have something deeply personal on the line, or because you have experienced the frustration of seeking transparency from institutions with considerably more resources and power than you have, remember this:

You are not alone.

A disappointing decision does not have to be the end of a story.

An unanswered question does not cease to matter because someone in authority declined to answer it.

And accountability does not require us to begin by assuming the worst about government.

It requires something much simpler:

Keep asking. Keep documenting. Keep comparing what we are told with what the record actually shows.

Meaningful change rarely happens because one person possesses all the power.

It happens when people become informed.

When records become accessible.

When experiences can be compared.

When evidence is preserved.

When journalists keep asking questions.

And when ordinary people decide that transparency is worth participating in.

That is something every reader can help accomplish.

Subscribe below to The Maine Mirror for free.

Share these investigations.

Read the source documents.

Engage respectfully online.

Send an article to someone who needs to see it.

Ask questions.

Challenge our reporting when the evidence warrants it.

And help put more eyes on the public record.

Every new reader makes important information harder to overlook.

Every share puts another set of eyes on the evidence.

Every informed conversation moves us another step toward the transparency, accountability, and justice our communities deserve.

If you feel discouraged by what you have seen from government institutions, you are not standing alone.

There are other people reading.

Other people asking questions.

Other people documenting.

Other people refusing to give up on the belief that public institutions can—and should—earn the public’s trust.

Together, we can keep the record visible.

Subscribe for free.

Share the reporting.

Join the conversation.

And keep asking for the transparency, accountability, and justice we all deserve.

The work continues.

And we continue it together.


Discover more from The Maine Mirror

Subscribe to get the latest posts sent to your email.

Leave a Reply

Discover more from The Maine Mirror

Subscribe now to keep reading and get access to the full archive.

Continue reading