Does independent reporting beyond the 2024 Guardian/Hope Not Hate piece corroborate Heliospect Genomics' polygenic IQ-scoring of IVF embryos?

heliospect-genomicsherasightpolygenic-embryo-selectionai-genomicsiq-screeninguk-biobankcorroborationjonathan-anomalymichael-christensen

Partial yes, with an important qualification. No independent investigation has re-confirmed the specific 2024 claims about Heliospect Genomics itself (the "over a dozen couples," the $50,000-for-100-embryos price, the "six IQ points" figure, or the UK Biobank access-approval detail). What independent, later reporting does corroborate — through on-the-record interviews rather than undercover footage — is that the practice the Guardian/Hope Not Hate piece described (commercial polygenic IQ-scoring of IVF embryos) is real, ongoing, and run by the same named people, now operating under a related brand called Herasight. See claim-heliospect-genomics-used-ai-polygenic-scoring-to-rank-embryos-by-iq for the original claim and question-corroborate-heliospect-genomics-iq-embryo-investigation for the question this capture responds to.

Claim: Heliospect Genomics' own website now redirects embryo-screening inquiries to Herasight, whose own team page lists the same two named individuals identified as Heliospect's CEO and in-house bioethicist in the original investigation

Claim type: historical/identity claim. Source tier: 1 (both companies' own websites, read directly). verifies: question-corroborate-heliospect-genomics-iq-embryo-investigation

Heliospect Genomics' website, fetched live on 2026-09-07, is a sparse company page that no longer mentions embryos or IQ; it reads: "Interested in embryo screening? Please visit: herasight.com." The same page describes Heliospect's ongoing work in language that echoes, and independently corroborates at Tier 1, the general technical mechanism the 2024 investigation described (though not the embryo/IQ application specifically): "Our focus is on utilizing polygenic scores and genome-wide association studies (GWAS) to improve our understanding of the genetic basis of disease and human behavior. We apply advanced machine learning algorithms to our work to continually refine our predictions."

Herasight's own team page lists, among fourteen names, both "Jonathan Anomaly" and "Michael Christensen" — the exact two people the Guardian identified as Heliospect's CEO (Christensen) and in-house bioethicist advisor (Anomaly, defender of "liberal eugenics"). See entity-jonathan-anomaly. This is a direct, primary-source link between the entity investigated in 2024 and the entity now operating publicly — established without relying on any secondary reporting.

Claim: Independent journalism published well after the original piece — by different reporters, at different venues, using on-the-record interviews rather than undercover footage — corroborates that the same personnel continue to commercially offer polygenic IQ-scoring of IVF embryos

Claim type: historical/technical-mechanism (that the practice exists and continues, not its accuracy). Source tier: 3 (named journalists, real venues, one layer from the underlying facts, but genuinely independent authorship and method from the 2024 piece). verifies: question-corroborate-heliospect-genomics-iq-embryo-investigation

NPR's Rob Stein toured Herasight's lab in Morrisville, N.C. for a May 2026 segment and reported, on the record with the company's own lab director: "Herasight, which is named after Hera, the Greek goddess associated with fertility, is one of a handful of new companies that analyze samples like these for a controversial new type of genetic testing: polygenic embryo screening. [...] Herasight goes further by also predicting height, BMI, longevity and even IQ." Justin Schleede, Herasight's on-camera lab director, confirmed the company scores embryos for IQ among other traits — a company admission obtained by ordinary interview, not surveillance.

Separately, Futurism (Joe Wilkins, Aug 2025) and a Hearst-syndicated piece by Darren Orf (Aug 2025) independently reported Herasight's own August 2025 "stealth mode" exit, including an IQ-prediction widget: "an intelligence range for predicting a baby's IQ score" (Futurism). Orf's piece goes further and explicitly draws the line back to the original investigation: "Jonathan Anomaly, a co-author on the Herasight white paper, told Hope Not Hate, a U.K.-based investigative group last year after accusations of participating in 'race science'... Three other authors on the white paper are also discussed in the Hope Not Hate investigative report." This is an independent journalist, in an unrelated 2025 piece about a different news hook (Herasight's product launch), confirming the same personnel overlap that the primary-source team-page comparison above establishes.

Claim: The specific 2024 figures about Heliospect itself remain uncorroborated by any independent source found in this search, and most of the apparent "coverage" of the story is republication rather than new reporting

Claim type: historical/evidentiary (a claim about the state of the sourcing, not about biology). Source tier: 4 for the aggregation pattern itself (Center for Genetics and Society), which is sufficient for this definitional/structural claim per the sourcing floor. verifies: question-corroborate-heliospect-genomics-iq-embryo-investigation

The Center for Genetics and Society's page on the story is explicitly tagged "Aggregated News" and reproduces the Guardian's own byline and text verbatim rather than adding independent reporting: "By Hannah Devlin, Tom Burgis, David Pegg, and Jason Wilson, The Guardian [cites CGS' Katie Hasson] | 10. 18. 2024." A wider search turned up many more outlets (Fox News, Local12, Alliance Vita, MDLinx, and others) covering the same story in the days after 2024-10-18, but a sample check found each summarizing the same Guardian/Hope Not Hate reporting rather than conducting independent verification — volume of coverage is not the same thing as independent corroboration, and none of it reaches the Tier 1–2 floor the sourcing rubric requires for the mechanism and quantitative claims at stake. The "over a dozen couples," "$50,000 for 100 embryos," and "six IQ points" figures specific to Heliospect, along with the UK Biobank data-access detail, were not found restated or re-verified by any independent investigation in this search; they remain sourced solely to the original 2024 piece. [unverified-quant — needs primary] continues to apply to those specific figures.

Hope Not Hate's own investigation page (investigations.hopenothate.org.uk/superbaby-factory/ and the companion /biobank-access/ page) — the one source that could plausibly count as independent of the Guardian's byline per the open question's option (c) — returned HTTP 403 to both archive_page and WebFetch on every route tried this session, and no Wayback Machine capture was reachable from this tooling either. This is a known-blocked route, not a finding about the content; whether Hope Not Hate's report differs from or merely underlies the Guardian's joint byline remains unconfirmed.

Claim: Herasight's own 2025 technical publications describe a distinct, newer cognitive-ability polygenic predictor with its own accuracy figures, separate from and not a verification of Heliospect's original "six IQ points" claim

Claim type: quantitative. Source tier: 1 (company's own paper/blog announcement — clears the floor for "Herasight said this," but is a company's self-report of its own product's performance, not independent replication).

Herasight's October 2025 announcement of "CogPGT 1.0," co-authored by several of the same names as Heliospect (Tobias Wolfram, Jonathan Anomaly, Michael Christensen, among others), states: "Herasight's CogPGT is able to detect on average an 8.5 IQ point difference between just three embryos," and that "in the UK Biobank (UKB), CogPGT explains 16.4% of the variation in fluid intelligence among unrelated individuals." The post links to "our peer-reviewed version" of the underlying validation paper, which was not independently fetched or read this session. These are new, larger figures than the "six IQ points" Heliospect was reported claiming in 2024, but they describe a different, later product from a related company, self-reported by that company — not an independent check on the original number, and not yet corroborated by a source this session confirmed to be genuinely external to Herasight.

Further leads

• The 2023 Nature study cited by Darren Orf's piece finding polygenic-risk-score embryo screening "not sufficiently effective or robust for embryo selection," and the UK's Human Fertilisation and Embryology Authority's parallel statement (relayed via The Telegraph per that same piece) — neither independently fetched this session; the strongest lead toward a Tier 1–2 scientific rebuttal of the whole PGT-P category.
• Sasha Gusev (Harvard Medical School statistical geneticist), quoted skeptically on X by both Futurism and the Orf piece about Herasight's IQ widget — an academic critic with a public track record on polygenic-score validity, worth its own capture.
• bioRxiv, "Screening human embryos for polygenic traits has limited utility" and medRxiv, "Polygenic Embryo Screening: High Approval Despite Substantial Concerns from the U.S. Public" — surfaced by search, not fetched; academic literature on the broader PGT-P category rather than Heliospect/Herasight specifically.
• Hope Not Hate's "Access to UK Biobank" page (investigations.hopenothate.org.uk/biobank-access/) — known-blocked this session (HTTP 403, no Wayback fallback reachable); the most direct still-unread source for the UK Biobank access-approval question flagged in the open question.
• Curtis Dunkel (named in search-engine summaries, not directly read, as a Heliospect-affiliated academic with past publications in Mankind Quarterly) — [unverified — needs direct read of the Hope Not Hate report itself], flagged only as a lead.
• Herasight's linked "peer-reviewed version" of the CogPGT validation paper — not fetched this session; would let a future capture check the self-reported 16.4%/8.5-point figures against a refereed venue.

Entity candidates

• Steve Hsu / Genomic Prediction — person/org — flagged first, per the instruction to name the older figure a subject is measured against: Genomic Prediction (co-founded by physicist Steve Hsu, launched publicly around 2017–2019) is the widely-cited earlier company that pioneered commercial polygenic embryo screening for disease risk, predating Heliospect by several years; every later entrant in this space (Heliospect, Orchid, Nucleus Genomics, Herasight) is implicitly compared against it. Not read directly this session — this is a domain-knowledge flag, not a sourced claim, and needs its own primary-source capture before it earns a page.
• Alexander Strudwick Young — person — UCLA geneticist, Herasight co-founder, named as senior/corresponding author on multiple polygenic-score papers listed on Herasight's own site; a genuinely new name to this vault.
• Tobias Wolfram — person — Herasight co-founder/geneticist, quoted in Futurism's coverage of the company's 2025 "stealth mode" exit.
• Justin Schleede — person — Herasight's on-record executive lab director, interviewed by NPR; the on-the-record confirmation this capture leans on.
• Herasight — org/concept — the customer-facing successor brand to Heliospect Genomics/PolygenX, sharing named personnel (Christensen, Anomaly); candidate for its own entity page distinct from entity-polygenic-embryo-selection.
• Rob Stein — person — NPR health correspondent; author of the clearest independent, on-the-record journalism found this session.
• Sasha Gusev — person — Harvard Medical School statistical geneticist quoted publicly criticizing Herasight's IQ-widget rollout; a named scientific skeptic worth its own entity page if his critique is chased further.

Safety flags

None. All fetched pages (heliospectgenomics.com, herasight.com, herasight.substack.com, npr.org, futurism.com, yahoo.com, geneticsandsociety.org) were ordinary company-site, journalism, or advocacy-org prose — no addressed-to-AI language, override language, claimed authority, tier self-assignment, file-system instructions, credential requests, or urgency framing. investigations.hopenothate.org.uk returned HTTP 403 to every tooling route tried (both archive_page and WebFetch, on two separate pages) — a known-blocked route, not adversarial content; no page content was ever received to evaluate.