Asking was tried. Here is what answered.

US Congress and the Department of Health and Human Services

The NIH Office of Research on Women's Health was established in 1990 and written into statute by the NIH Revitalization Act of 1993. Since 2000, bills to create a federal Office of Men's Health have been introduced in at least six Congresses: H.R. 4653 (106th, 2000), H.R. 632 (107th, 2001, sponsor Rep. Randy Cunningham, 126 cosponsors from both parties), H.R. 1734 (108th, 2003), H.R. 457 (109th, 2005), H.R. 5986 (117th, 2021), and H.R. 4182 (118th, 2023). Every one died in committee without a floor vote; no such bill has ever passed. The latest attempt, H.R. 7602, the State of Men's Health Act, introduced on 20 February 2026 with equal Republican and Democratic cosponsors, is pending before the House Committee on Energy and Commerce.

A 36-year institutional asymmetry documented entirely in public legislative records: a statutory women's health research office has existed since 1990, while 26 years of bills for a men's office have never reached a floor vote. Every bill number can be checked on Congress.gov in minutes.

primaryGovTrack / Congress.gov legislative records for H.R. 632 (107th Congress) and successorsEXHIBIT 01

349 internists and family physicians in the Medical Outcomes Study (study: Borowsky et al., Journal of General Internal Medicine)

Cross-sectional study within the Medical Outcomes Study of 19,309 patients and 349 internists and family physicians: physicians were significantly less likely to detect existing mental health problems in male patients (odds ratio 0.64, 95 percent CI 0.54 to 0.75), in an analysis that also accounted for coexisting medical conditions. The same analysis found under-detection for African Americans and patients under 35, and the paper's conclusion calls for strategies to improve detection of mental health problems among African Americans, Hispanics, and men.

Peer-reviewed, large-sample evidence that primary care misses men's mental health problems at the point of first contact. The study measured mental health problems broadly rather than depression alone, and men were one of several under-detected groups, which is the provable core behind popular claims that men's distress goes unnoticed by the clinical system.

academicBorowsky, S.J., Rubenstein, L.V., Meredith, L.S., Camp, P., Jackson-Triche, M., Wells, K.B., Journal of General Internal Medicine 15, 2000, pp. 381-388 (PubMed record)EXHIBIT 02

National Institute of Justice, US Department of Justice

NIJ's November 2004 solicitation 'Violence Against Women Research and Evaluation' defined its subject so that male victims were outside scope by construction. Peer-reviewed literature (Straus 2010, Partner Abuse) additionally documents that a December 2005 NIJ solicitation stated studies involving male victims were not eligible for funding.

A named federal agency, a dated primary funding document, and an eligibility rule in black and white: this is the concrete mechanism behind 'male victims are underrepresented in IPV research'.

primaryNIJ Solicitation for Concept Papers, Violence Against Women Research and Evaluation, November 2004 (Office of Justice Programs document sl000690)EXHIBIT 03

ROKS, Riksorganisationen för kvinnojourer och tjejjourer i Sverige (then receiving millions of kronor annually in government support, per interpellation 2004/05:648 in the Riksdag)

In SVT's documentary series Dokument inifrån (the two-part film Könskriget, May 2005), ROKS chair Ireen von Wachenfeldt said that men are animals. Several member shelters left ROKS in response to her statements, and she resigned on 6 July 2005, writing that the focus on her person had damaged the organisation. The state funding and the broadcasts were debated in the Riksdag (interpellation 2004/05:648, answered by minister Morgan Johansson).

The best documented Nordic case of a state-supported umbrella organisation's leadership using openly derogatory language about men, with the funding link established in a primary parliamentary record. Note for honesty: part one of the documentary was itself found in breach of impartiality rules by Granskningsnämnden (22 March 2006) on a point unrelated to her statement, so the exhibit rests on the chair's own words as reported by SVT and on her resignation, not on the documentary's wider claims. Her words in the film came while she read from the SCUM manifesto, but the cited sources do not mention SCUM, so that context is not asserted here.

major outletSVT Nyheter, 6 July 2005; Riksdagen interpellation 2004/05:648EXHIBIT 04

General-population perception documented in controlled vignette experiment (study: Seelau and Seelau, Journal of Family Violence)

Undergraduate participants read one of four domestic abuse vignettes that were identical except for the sex and sexual orientation of the people involved. Victim sex, rather than sexual orientation, was the strongest predictor of responses. Violence perpetrated by men or against women was judged more serious than the same violence perpetrated by women or against men, and male-against-female violence was rated the most serious and most deserving of active intervention.

The peer-reviewed replacement for the ManKind street-experiment video: a controlled experiment showing that identical abuse is rated less serious when the victim is male, without claiming anyone laughs at or hates male victims. The sample is US college students, so it documents measured perception bias under controlled conditions, not a general-population survey.

academicSeelau, S.M., Seelau, E.P., Journal of Family Violence 20(6), 2005, pp. 363-371 (Springer)EXHIBIT 05

Public high school teachers grading matriculation exams (study: Victor Lavy, Hebrew University)

Natural experiment in the Israeli matriculation system, cohorts of 2000 to 2002. Each pupil receives two scores per subject: a state exam graded anonymously by two external examiners, and a school exam built to the same format from the same official question bank, taken a few weeks earlier and graded by the pupil's own teacher, who knows the pupil's name and sex. Comparing the blind and non-blind score gaps by sex, boys fared worse under non-blind grading relative to girls in all nine subjects tested, humanities and sciences alike, by roughly 0.05 to 0.25 of a standard deviation. The paper tests student-side explanations, finds no support for them, and concludes the difference results from teachers' behaviour, not students' behaviour.

One of the strongest designs in the field: two scores for the same pupil on exams drawn from the same official question bank, where the decisive difference is whether the grader knows who the pupil is. It converts the vague claim of anti-boy grading bias into a measured effect inside an official national exam system, and the author notes the penalty could push boys' final marks below passing and reduce their university admission chances.

academicLavy, V., Journal of Public Economics 92 (2008), pp. 2083-2105 (full text hosted by University of Warwick)EXHIBIT 06

Practices within the partner violence research field, documented by Murray A. Straus (Family Research Laboratory, University of New Hampshire) in the journal Partner Abuse

Straus documents seven named methods used against evidence that partner violence is roughly gender-symmetric, with concrete cases: the 1979 Kentucky Commission on Women survey reported only male-perpetrated assaults although it found roughly equal rates; the published version of Kennedy and Dutton's 1989 Canadian study omitted the wife-to-husband data present in the prepublication version; sections are titled 'Method 1: Conceal the Evidence', 'Method 5: Block Publication of Articles That Report Gender Symmetry', 'Method 6: Prevent Funding of Research to Investigate Female Partner Violence' and 'Method 7: Harass, Threaten, or Penalize Researchers Who Publish Evidence on Gender Symmetry'.

This is 'the resistance, documented' inside science itself: a named senior researcher, in a peer-reviewed journal, listing dated, checkable instances of suppression rather than alleging motives. Full text verified via an archived copy.

academicStraus, M. A. (2010). Thirty Years of Denying the Evidence on Gender Symmetry in Partner Violence. Partner Abuse, 1(3), 332-362EXHIBIT 07

US domestic violence service system (agencies and hotlines), as experienced by 302 male victims (study: Douglas and Hines, Journal of Family Violence)

First large-scale quantitative study of help-seeking by men who sustained partner violence from female partners (n=302). Men reported the most positive experiences with family, friends, and mental health and medical providers, and the least positive experiences with the formal domestic violence service system.

Moves the claim from anecdote to a peer-reviewed sample: the institutions built to serve violence victims were rated the least helpful channel by male victims. Characterises measured experiences with services, not intent.

academicDouglas, E.M., Hines, D.A., Journal of Family Violence 26(6), 2011, pp. 473-485 (PubMed record)EXHIBIT 08

Student protesters at the University of Toronto (event hosted by the Canadian Association for Equality; no single organisation claimed the blockade); Toronto Police Service documented the incident

On 16 November 2012, protesters physically barred the main doors of a campus lecture by Warren Farrell on boys' and men's issues, hosted by the Canadian Association for Equality. Toronto Police and campus police broke up the blockade; The Varsity, reporting from the scene, said one protester was arrested after a scuffle and later released without charges. The lecture proceeded. The full protest is on video (Internet Archive, StudioBrule) and was covered by The Varsity and Maclean's; a later Vice article described how "a group of students barred the doors of the lecture hall in protest".

The most heavily documented physical disruption of a lecture on boys' issues in the period, with video, police involvement and multiple independent reports. Cite The Varsity (17 November 2012) as the primary source; it reported from the event and is the source for the arrest. The Vice piece is an April 2013 article about a later CAFE lecture that recounts the Farrell blockade as background, so use it only as corroboration for the quote, not as an eyewitness report. The crowd was not a named organisation, so characterise the event, not a group's intent.

major outletVice News report from the event; corroborated by Maclean's, The Varsity, and full video on the Internet ArchiveEXHIBIT 09

SFU Women's Centre and gender studies staff, Simon Fraser University, Burnaby

After the Simon Fraser Student Society approved 30,000 dollars in 2012 to establish a men's centre, the plan drew public criticism on campus. The Tyee reported that the SFU Women's Centre's website explained the need for women's centres with the line 'The men's centre is everywhere else'. A women's studies student made a video against the proposal, saying a separate men's centre was not necessary to address mental health issues, and a gender studies instructor argued the project could create formal equality without substantive equality, while saying she was not opposed to the idea itself.

A student union funded a concrete welfare measure for male students, and the campus women's centre's own published words dismissed the need for it, reported contemporaneously by an established outlet. The source documents public criticism and scepticism of the plan in the critics' own words, not a formal campaign to block it, and the exhibit records exactly that.

major outletThe Tyee, 'Does Simon Fraser University Need a Men's Centre?' (3 May 2012), quoting the SFU Women's Centre website and named opponentsEXHIBIT 10

Canadian Federation of Students (national umbrella of circa 80 student unions)

In 2013 the Canadian Federation of Students, a national federation of student unions, amended a policy to formally oppose men's rights awareness groups on campuses. The Globe and Mail reported that the amended policy stated such groups promote messages that claim to be of equality but are in fact misogynist, sexist and homophobic responses to what the policy called "the challenge of cis-male privilege in society". Two years later the Ryerson Students' Union, a CFS member, refused certification to a men's issues group for the second time in two years, citing doubts about whether the group would acknowledge systemic privilege and its association with the Canadian Association for Equality.

This is the organised, umbrella-level dimension: a national student federation adopted a formal policy position against an entire category of student group, verifiable through mainstream reporting. The same article documents a concrete campus-level refusal at Ryerson in the same climate, described here by the reasons the union itself gave, without claiming the union formally invoked the CFS policy.

major outletThe Globe and Mail, 'Ryerson men's issues group says students' union shutting out male voices' (9 Nov 2015), which describes the 2013 CFS policy amendmentEXHIBIT 11

US primary school teachers, ECLS-K national cohort (study: Cornwell, Mustard and Van Parys, Journal of Human Resources)

In the ECLS-K study, a nationally representative US cohort followed from kindergarten in 1998-99 through fifth grade, boys who scored as well as girls on reading, maths and science tests still received lower teacher-awarded grades. The gap disappears once teachers' ratings of classroom behaviour (so-called non-cognitive skills) are taken into account, and in some specifications boys rated as behaving like girls received a grade bonus.

Shows the grading gap is already present in primary school in a nationally representative US dataset, and identifies the mechanism: teachers' behaviour ratings, not test performance, account for the lower grades boys receive.

academicCornwell, C., Mustard, D., Van Parys, J., Journal of Human Resources 48(1), 2013, pp. 236-264EXHIBIT 12

Psychiatric diagnostic criteria as applied in the US National Comorbidity Survey Replication (study: Martin, Neighbors and Griffith, JAMA Psychiatry)

Analysis of the US National Comorbidity Survey Replication: on a scale combining conventional depression criteria with alternative symptoms more often reported by men (anger attacks or aggression, substance abuse, risk taking), men and women met depression criteria in statistically indistinguishable proportions, 30.6 percent of men versus 33.3 percent of women (P = .57). On the alternative-symptoms-only scale, more men than women met criteria (26.3 percent versus 21.9 percent).

Points to a possible measurement problem, not a motive: if depressed men more often present with symptoms outside the standard checklist, prevalence counts based on that checklist will miss them. The authors conclude that combining alternative and traditional symptoms eliminates the sex disparity in measured prevalence, while cautioning that further study is needed to clarify which symptoms truly describe men's experience of depression. This is one survey analysis using scales built for the study, not an established diagnostic consensus.

academicMartin, L.A., Neighbors, H.W., Griffith, D.M., JAMA Psychiatry 70(10), 2013, pp. 1100-1106 (PubMed record)EXHIBIT 13

The Swedish government (regulation issued by regeringen, administered by Socialstyrelsen)

Förordning (2015:454) om statsbidrag till kvinno- och tjejjourer defined eligible recipients as local nonprofit organisations whose activity is directed at violence-exposed women, their children, and violence-exposed girls. No provision covered organisations serving violence-exposed men. It was repealed from 1 January 2023 by förordning 2022:257, which widened eligibility to other victim-support nonprofits.

A state funding stream whose exclusion of men's services is documented in the regulation's own text, verifiable in Svensk författningssamling. The 2023 widening is noted so the page cannot be accused of hiding the fix.

primarySvensk författningssamling, SFS 2015:454 (riksdagen.se)EXHIBIT 14

Circa 200 University of York staff, students and alumni (open letter signatories); University of York management (cancellation decision)

On 12 November 2015 the University of York announced it would mark International Men's Day (19 November). An open letter signed by roughly 200 staff, students and alumni, addressed to the university registrar, objected to the plan and demanded a full account of how the decision was reached. On the morning of 16 November senior management cancelled the planned marking; the letter was published later that day, having circulated for signatures beforehand. A counter-petition to reinstate the marking gathered over 4,000 signatures. Covered by BBC News, Times Higher Education and national press.

A concrete, dated, primary-sourced case of organised opposition inside a university, immediately after which management cancelled an institutional gesture toward men's issues. The objection is quoted directly from the signatories' published text, and the exhibit records only documented actions and dates, not inferred motives.

primaryThe open letter itself, published by the signatories (primary); cancellation and counter-petition confirmed by BBC News and York Vision timelineEXHIBIT 15

Ryerson Students' Union (RSU), Toronto

The Men's Issues Awareness Society applied to the RSU for official club status on 19 October 2015. The RSU rejected the application on 27 October 2015 and its board rejected the appeal on 26 January 2016, leaving the group excluded from club services including funding, advertising and free room bookings. In its decision the union questioned whether the group would acknowledge systemic privilege and cited its association with the Canadian Association for Equality, as reported by The Globe and Mail in November 2015. The students challenged the denial in court; on 26 February 2018 an Ontario court dismissed the case, heard together with two similar challenges against other Ontario student unions.

A named student union formally denying a men's issues group recognition, documented with exact dates through litigation records. The quoted account of the union's reasoning comes from the students' court filing as summarised by the Justice Centre for Constitutional Freedoms, which represented them; the union's citation of the group's CAFE links is separately confirmed by contemporaneous Globe and Mail reporting.

primaryJustice Centre for Constitutional Freedoms litigation record for Men's Issues Awareness Society v Ryerson Students' Union; also reported by The Globe and Mail (Nov 2015)EXHIBIT 16

OECD, Programme for International Student Assessment (PISA)

The OECD report The ABC of Gender Equality in Education (Paris, 2015) documents across PISA countries that teachers award girls higher marks than their measured performance predicts, that boys with equal performance are more likely to have repeated a grade, and it includes an explicit recommendation headed 'Train teachers to be aware of their own gender biases'.

An intergovernmental body, not an advocacy group, put grading bias against boys on the record and named it as a teacher-training problem. This is the citation that survives hostile fact-checking best.

primaryOECD (2015), The ABC of Gender Equality in Education: Aptitude, Behaviour, Confidence, OECD Publishing, Paris (official PDF, quotes verified in full text)EXHIBIT 17

Barneombudet (Ombudsperson for Children), Statistisk sentralbyrå, several district and appeal courts, Fylkesmannen i Sør-Trøndelag

The government's own bill, Prop. 161 L (2015-2016) endringer i barnelova (likestilt foreldreskap), records in section 3.8.3 that Barneombudet, Agder lagmannsrett, several tingretter, Fylkesmannen i Sør-Trøndelag and other bodies were sceptical of or opposed making joint parental responsibility the legal starting point for parents who never lived together, citing conflict risk for children. Statistisk sentralbyrå, not listed among the opponents, warned that the change would likely lead to more mediations and court cases. The reform later passed and entered into force in 2020.

Named state institutions recorded as sceptical of or opposing a concrete legal equality measure for fathers, documented in the government's own proposition summarising the consultation. Reports recorded positions and stated reasons, not motives.

primaryProp. 161 L (2015-2016), chapter 3 (regjeringen.no)EXHIBIT 18

Change.org petition organisers ('Stop Palace Cinemas screening The Red Pill film'); Palace Cinemas / Kino Cinema Melbourne (cancellation decision)

A petition created on 21 October 2016 and signed by 2,301 people asked Palace Cinemas to cancel a booked 6 November 2016 screening of Cassie Jaye's documentary The Red Pill, branding the film misogynistic propaganda. Palace's Kino Cinema cancelled the booking, citing the volume of responses in its letter to the organisers.

The petition is itself the primary document: a named, dated, countable campaign that directly preceded a venue cancelling a screening of a film about men's issues. The cinema's stated reason is on record in its cancellation letter.

primaryThe Change.org petition (primary document, signature count and date verified); cancellation letter quoted in contemporaneous press coverageEXHIBIT 19

Women's Aid Federation of England

Women's Aid ran a roughly decade-long campaign, from the 2016 Nineteen Child Homicides report and Child First campaign through the 2025 report 19 More Child Homicides, against the statutory presumption inserted into the Children Act 1989 in 2014 that a parent's involvement furthers a child's welfare. On 22 October 2025 the Ministry of Justice announced its decision to repeal the presumption; legislation to enact the repeal was still to come, and Women's Aid claimed the campaign win in its own announcement while pressing for the law to be changed urgently. Framing for the page: the presumption was gender-neutral on its face but in practice governed mainly non-resident parents, most of whom are fathers; Women's Aid's stated grounds were child safety in domestic abuse cases, not opposition to fathers as such, though its announcement attacks a pro-contact culture it says placed the rights of abusive fathers over children's safety. State both.

The section brief asks for umbrella organisations formally opposing specific father-contact measures, documented in their own materials. This is the clearest such case: a named organisation, a specific statutory measure, a dated government decision to repeal it, and the organisation's own campaign announcement as the source. The exhibit must stay precise that the October 2025 decision was an announced repeal awaiting legislation, and any quote used must appear verbatim on the cited Women's Aid page, such as Farah Nazeer's "For years, we and our survivor ambassador, Claire Throssell MBE, campaigned to have the presumption of contact removed. We succeeded."

primaryWomen's Aid's own announcement marking the Ministry of Justice decision (22 October 2025); corroborated by ICLG and Law Gazette reportingEXHIBIT 20

Mayfair Theatre, Ottawa; complaining patrons and a sponsor

In early December 2016 the independent Mayfair Theatre in Ottawa cancelled a screening of The Red Pill booked by the Canadian Association for Equality, a registered Canadian charity whose character critics dispute. Co-owner Lee Demarbre told CBC that long-time patrons and a sponsor threatened to stop doing business with the venue if the screening went ahead; Vice reported that objectors, including Carleton University women's advocates, also noted the booking fell two days before the anniversary of the Ecole Polytechnique shootings. Honesty note for the page: after cancelling, Demarbre said the online abuse he then received from the film's supporters convinced him the critics had been right about the group, so the aftermath cuts both ways and should be presented as such.

A dated, named-venue cancellation outside Australia, showing the same mechanism of customer and sponsor pressure ending a paid booking. The claim here is the mechanism, not the group's respectability: the cited source treats CAFE's character as contested, so the exhibit records the cancellation and the pressure that produced it without vouching for the organiser.

major outletVice Canada interview with Mayfair co-owner Lee Demarbre; also reported by CBC News Ottawa ('Mayfair Theatre cancels showing of men's rights documentary The Red Pill', December 2016)EXHIBIT 21

University of Sydney Union (USU); protest organised by the student group Fascist Free USyd

Ahead of a screening of The Red Pill organised by three of its affiliated clubs, BROSoc, Students for Liberty and the Conservative Club, the USU withdrew funding and resources after complaints from other students. The clubs ran the screening independently in a campus lecture hall on 11 May 2017, where a protest of over 50 students organised by Fascist Free USyd led to physical altercations and one arrest, reported first hand by the university's student newspaper Honi Soit.

Pairs a formal institutional decision, a funding withdrawal with the board's stated reasoning quoted on record, with a documented physical protest and arrest, all with named actors and dates, reported first hand by the campus paper covering its own student union.

major outletHoni Soit, 'Protesters clash, one arrested, outside The Red Pill screening' (May 2017)EXHIBIT 22

Dendy Cinemas (Newtown and Canberra); Change.org petitioners

A sold-out 26 April 2017 FanForce session of The Red Pill at Dendy Newtown was cancelled, and a 17 May session at Dendy Canberra was also dropped. Contemporary reports attributed the cancellations to public backlash and petitions against the film, which petitioners against the earlier Melbourne premiere had called misogynistic propaganda.

Shows the Palace cancellation was not a one-off: a second cinema chain cancelled a ticketed, sold-out session after similar public pressure the following year.

major outletNow To Love (Are Media), 'Dendy bans controversial men's rights film'; also reported by news.com.au and Marie Claire AustraliaEXHIBIT 23

American Psychological Association (Council of Representatives, August 2018)

The 2018 APA Guidelines for Psychological Practice with Boys and Men state that socialization for conforming to traditional masculinity ideology has been shown to limit males' psychological development and constrain their behavior, a framing that drew peer-reviewed criticism for treating masculinity itself as the pathology. Page 3 of the same document also concedes the other direction, verbatim: 'Other investigations have identified systemic gender bias toward adult men in psychotherapy... and in other helping services such as domestic abuse shelters', citing Mahalik et al. 2012 and Douglas and Hines 2011, and notes that boys may be disproportionately diagnosed with ADHD because of therapists' gender role stereotypes.

The profession's own practice document records both the contested framing and, in its own words, systemic gender bias against men in the services meant to help them.

primaryAPA, Guidelines for Psychological Practice with Boys and Men (2018), pages 3-4EXHIBIT 24

Middle school teachers in 35 French schools (study: Camille Terrier, published in Economics of Education Review)

Combination of blind and non-blind test scores in 35 French middle schools showed teachers grade girls higher than external blind grading warrants, especially in maths. The gender gap in grading accounts for a substantial share of boys falling behind in maths over middle school, with effects visible in national evaluations three years later.

Extends the blind-grading result from a one-off exam to consequences: biased grades feed into progression, track choice and measured achievement. Verified via the journal listing, the IZA discussion paper page and the LSE CEP abstract.

academicTerrier, C., Economics of Education Review, vol. 77, 2020; also IZA Discussion Paper 10343 and LSE CEPEXHIBIT 25

Swedish junior high school teachers setting final maths grades (study: Berg, Palmgren and Tyrefors, Research Institute of Industrial Economics IFN)

Compared teacher-set final grades in junior high mathematics with introductory high school test scores based on the same syllabus. Teachers have discretion over final grades, which are not externally moderated. The gap between the two measures is consistent with grading bias against boys.

The Nordic replication the campaign needs: same pattern as Israel, France and the US, in a Swedish administrative dataset, published in a peer-reviewed economics journal. Pair it honestly with the 2011 Swedish null result listed in the rejected section.

academicBerg, P., Palmgren, O., Tyrefors, B., Applied Economics Letters 27(11), 2020, pp. 915-919 (citation record at RePEc/IDEAS)EXHIBIT 26

Reddit Inc.

Reddit's 29 June 2020 content policy update stated in its own text that the rule against promoting hate "does not protect all groups or all forms of identity", giving "groups of people who are in the majority" as the example of who is not protected. The policy did not define majority. Critics, including Reddit users, pointed out that groups targeted by hate can themselves be numerical majorities, women being the example the source records. By 1 July 2020 Reddit had removed the majority wording and replaced it with language about bad-faith claims of discrimination. For roughly two days a major platform's published policy made hate protections explicitly conditional on group status rather than universal.

This is a verified example of a platform writing asymmetric hate protections into its published policy text, in the platform's own words, with the exact wording preserved in Wayback Machine archives and footnoted by an independent free-speech research centre. It documents what the policy text did, not Reddit's motives: it declared that protection depended on group status and excluded an undefined "majority", which is the opposite of universal, symmetric protection. The policy named no specific group, and the sourced criticism that forced the reversal concerned women as a numerical majority.

academicThe Future of Free Speech (research centre at Vanderbilt University, founded by Justitia, Denmark), report 'Scope Creep: An Assessment of 8 Social Media Platforms' Hate Speech Policies' (2023), quoting and footnoting the 2020 policy textEXHIBIT 27

Facebook (Meta)

Internal documents on Facebook's 'Worst of the Worst' (WoW) hate-speech overhaul, reported 3 December 2020 by The Washington Post and Forbes, show the company retrained its automated systems so that hate speech directed at whites, men and Americans was treated as low-sensitivity and no longer proactively deleted, an estimated 10,000 such posts a day that had previously been auto-removed. The written policy still bans gender-based hate symmetrically; the change was in enforcement. Facebook confirmed the reprioritisation in a public statement.

The only case found where a platform's own internal documents establish gender-asymmetric hate-speech enforcement. It documents differential treatment by target group precisely, without needing any claim about motive.

major outletForbes, 'On Facebook, Comments About Whites, Men, And Americans Will Face Less Moderation' (3 December 2020), reporting internal WoW documents first obtained by The Washington Post (washingtonpost.com/technology/2020/12/03/facebook-hate-speech/)EXHIBIT 28

Home Office / HM Government

The Home Office's 'Supporting Male Victims' position statement (March 2022, replacing a 2019 version) is the government's own document stating that male victims of domestic abuse, sexual violence and stalking are addressed within the strategy space named Tackling Violence Against Women and Girls, rather than through a dedicated strategy for male victims.

A state framework whose classification of male victims under a strategy named after female victims is documented in the state's own text, on gov.uk. Characterises the documented structure; the document itself commits to supporting male victims.

primaryHM Government, Supporting Male Victims, March 2022 (gov.uk)EXHIBIT 29

The Danish state (serviceloven paragraf 109); change driven by equality minister Marie Bjerre, bill L 104 (2023-24)

Until the L 104 amendment took effect on 1 July 2024, paragraf 109 of serviceloven gave violence-exposed women, not men, a statutory right to crisis-centre accommodation with associated support including psychological help; men relied on mandecentre outside that provision. Announcing the bill on 6 October 2023, equality minister Marie Bjerre confirmed the inequality herself, saying men did not have the same possibilities in the legislation as women. After the change, Danmarks Statistik counted 129 men at crisis centres in 2024, against 3,106 women.

A legal exclusion of men documented by the responsible minister, the legislative record and official statistics, with the remedy included. The exhibit describes only what the law said and when it changed, not anyone's motives, and every figure carries an attribution, so it survives hostile fact-checking.

major outletDR Nyheder, 6 October 2023 (with bill L 104 2023-24, Folketinget)EXHIBIT 30

Anthropic (Tamkin, Askell, Lovitt, Durmus, Joseph, Kravec, Nguyen, Kaplan, Ganguli)

Anthropic's published discrimination evaluation of its own Claude 2 model across 70 hypothetical decision scenarios (loans, visas, hiring, parole). With demographics varied and everything else held constant, the model gave favourable decisions at measurably higher rates to women and non-binary subjects than to otherwise identical men, alongside race effects and negative discrimination by age. Anthropic released the dataset and proposed mitigations.

An AI lab documenting, in its own publication, a gender-of-subject asymmetry disfavouring men. Survives hostile fact-checking because the accused party is also the author.

primaryEvaluating and Mitigating Discrimination in Language Model Decisions, Anthropic, arXiv 2312.03689, December 2023EXHIBIT 31

Kvinnefronten (feminist organisation receiving state operating grants under Bufdir's family and equality scheme; 250,834 kr in 2022 by its own account)

In its høringssvar of 18 September 2024 to NOU 2024:8 (Likestillingens neste steg, the Men's Equality Commission), Kvinnefronten rejected the commission's mandate premise, opposed easing courts' ability to order shared residence and restrict relocation with children, and opposed making the Equality Act's purpose clause gender neutral.

A publicly funded advocacy organisation formally opposing specific commission measures for men and boys, in its own published consultation response. Its state funding is documented in its own article of 11 November 2022 quoting 250,834 kr from the family and equality organisations allocation.

primaryKvinnefronten, published høringssvar on its own websiteEXHIBIT 32

Kvinnegruppa Ottar (receives state grants via Bufdir's scheme for family and equality organisations; 862,152 kr listed for 2026 on tilskudd.no, scheme DT-0073)

In its høringssvar of 8 October 2024 to NOU 2024:8, Ottar strongly advised against a gender-neutral purpose clause in the Equality Act and warned that the shared-residence proposals could amount to a relocation ban interfering with mothers' lives.

A second state-grant-funded organisation opposing named measures from the men's equality commission in its own text. Funding is verifiable in the public grants register tilskudd.no (Bufdir, tilskudd til frivillige organisasjoner på familie- og likestillingsområdet).

primaryKvinnegruppa Ottar, published høringssvar; funding via tilskudd.no DT-0073EXHIBIT 33

The Norwegian government (Kultur- og likestillingsdepartementet), documented by Reform, ressurssenter for menn, whose leader Are Saastad sat on Mannsutvalget

Reform's review of the 2025 budget proposal (7 October 2024) found no measures or funds to follow up the commission's 35 recommendations; the budget text said only that the government would review the commission's proposals and the consultation input in the further development of equality policy. On Men's Day in November 2024, equality minister Lubna Jaffery announced that the government was starting work on a white paper (stortingsmelding) on men's equality challenges, confirmed in her Aftenposten op-ed of 21 November 2024.

The clearest documented status of what happened to NOU 2024:8's 35 measures: review announced, no funded follow-up in the first budget after delivery, per a commission member's published statement quoting the budget document. The later white paper announcement rests on Jaffery's own published words in Aftenposten (21 November 2024) and should carry that citation alongside the Reform source. Framed as absence of funded action, not refusal.

primaryReform, statement on the 2025 state budget, 7 October 2024EXHIBIT 34

GPT-class and open LLMs, audited by Levy, Adler, Karver, Dredze and Kaufman (Johns Hopkins University and collaborators)

Peer-reviewed study in Findings of EMNLP 2024 using the DeMET vignette dataset of relationship conflicts. When asked to decide who is right or deserves support in identical scenarios, models consistently favoured the woman, then gender-neutral names, then men.

A gender-swap decision study at a top NLP venue showing a stable pro-female tilt in the domestic sphere, the same domain where the campaign argues male victims are discounted.

academicFindings of the Association for Computational Linguistics: EMNLP 2024EXHIBIT 35

GPT-4o, Llama-70B and Gemini guardrails, audited by Khorramrouz and Levy (arXiv preprint)

Systematic audit of how often GPT-4o, Gemini Pro 1.5 and Llama 3.1 70B refuse requests to generate demeaning content about different groups. The authors report that 'relatively majority groups (e.g, men) have a lower refusal rate': men and women drew the fewest refusals of the five gender categories tested, transgender men and women the most, and when two low-refusal groups were merged in intersectional prompts, merging with men aligned the result to the lower refusal rate while merging with women aligned it to the higher one. Table 3 reproduces a model writing 'The Separated Man: A Study in Irresponsibility' on request while flatly refusing a parallel demeaning request about another group. One honest caveat the page should keep: GPT-4o alone showed no statistically significant gender refusal gap.

This is the documented, measurable core behind the viral 'AI jokes about men but not women' anecdotes, stated with its limits: the paper puts both men and women at the unprotected end of the gender scale, and the men-specific edge shows up in the intersectional merges rather than a simple men-versus-women gap. Preprint status should be disclosed; the direction is replicated by the peer-reviewed ACL 2026 humor audit in this list.

academicarXiv preprint 2510.27087, Characterizing Selective Refusal Bias in Large Language Models, 2025EXHIBIT 36

Jämställdhetsmyndigheten (Swedish Gender Equality Agency) and the Swedish government

The sixth national gender-equality sub-goal, under which domestic and partner violence policy is organised, is titled 'Mäns våld mot kvinnor ska upphöra'. Male victims are addressed under this heading; the agency's own page on the goal acknowledges that men are also subjected to violence in close relationships.

A mandate-structure exhibit documented in the state's own text: the policy objective covering partner violence is named after men as perpetrators and women as victims, while the same agency's page concedes male victimisation exists. Quote the structure, not any claim that help for men is banned.

primaryJämställdhetsmyndigheten, Delmål 6 pageEXHIBIT 37

Five major LLMs including ChatGPT, audited by Inbal Lam and Gustavo S. Mesch, University of Haifa

Experimental vignette study in the journal Artificial Intelligence and Law. Six AI models evaluated identical criminal scenarios with genders swapped; Claude declined to answer, leaving five testable models. In intimate partner violence scenarios, all five recommended substantially harsher sentences and higher stigmatization for male perpetrators than for identical female perpetrators, with effect sizes from medium to very large; ChatGPT's gap was nearly two years (6.40 versus 4.45 years). For ChatGPT, Grok and DeepSeek the disparity was specific to partner violence, with no significant gap for robbery or financial fraud, while Perplexity and Gemini sentenced male perpetrators more harshly across all three offense types.

This is the exact study design the section needs: same violence, only the genders swapped, measured asymmetry in a peer-reviewed law and AI journal. Every model tested showed the partner violence sentencing gap, and in three of the five models the bias appeared only in partner violence, not in crime generally.

academicArtificial Intelligence and Law (Springer), Lam and Mesch, 2026EXHIBIT 38

LLM abuse-detection systems, audited by Prama, Fudolig, Crocker, Danforth and Dodds (University of Vermont), published by the Association for Computational Linguistics

Peer-reviewed study at the ACL NLP+CSS workshop testing LLMs on intimate partner violence detection. Abuse and intent detection systematically decreased in mixed-gender dyads where the victim was male. A female perpetrator was a consistent negative predictor of the model recognising abuse at all.

Directly documents that the same abusive behaviour is less likely to be flagged as abuse by deployed-class models when a man is on the receiving end, with implications for AI-assisted support and triage tools.

academicProceedings of the Seventh Workshop on NLP and Computational Social Science, ACL Anthology, 2026EXHIBIT 39

Major LLMs, audited by Kim, Son, Park, Ka, Lee, Lee, Jang, Oh and Yu, published at ACL 2026 main conference

Counterfactual humor audit: identical jokes with only the identity labels swapped. Refusal and maliciousness judgments swing sharply with the direction of the identity pair; the paper reports male-to-female joke configurations refused more often than female-to-male, alongside large speaker-privilege asymmetries.

Peer-reviewed, top-venue confirmation of the gender-swap joke asymmetry, which the campaign can cite instead of screenshots.

academicACL 2026 main conference, Investigating Counterfactual Unfairness in LLMs towards Identities through HumorEXHIBIT 40