Exclusive: OpenAI, Anthropic meet with House Homeland Security behind closed doors on cyber threats

Axios · 🇨🇳 public China · Apr 28, 2026 · 726 words · By Sam Sabin; Ashley Gold

AI National Security Risks US-China AI Competition Government Regulation of AI

headphones Listen to the eFinder podcast briefing

Generate a natural audio summary of this story

Daily briefing

What to know about AI National Security Risks

OpenAI and Anthropic provided classified briefings to the House Homeland Security Committee regarding the cybersecurity risks associated with their latest AI models. The article also details a separate briefing on 'jailbroken' models and the resulting calls from lawmakers for increased regulation to ensure public safety.

Propaganda risk 30%

Claims checked 12

Techniques found 2

Topics 3

Coverage spectrum

Coverage gap: Low Left coverage

Left0%

Center75%

Right25%

4 sources compared across this story cluster. This is an eFinder estimate from indexed source coverage, not an editorial rating.

What happened

OpenAI, Anthropic brief House Homeland Security on AI cyber threats Skip to main content Apr 28, 2026 - Technology Exclusive: OpenAI, Anthropic meet with House committee over advanced cyber models Sam Sabin, Ashley Gold email (opens in new window) sms (opens…

Why it matters

Add Axios on Google House Homeland Security Chair Andrew Garbarino during a hearing in March.

Common ground

Photo: Tom Williams/CQ-Roll Call via Getty Images OpenAI and Anthropic briefed House Homeland Security Committee staff on their new cyber-capable AI models and their implications for cybersecurity, Axios has learned.

Perspective signals

The tension in the story is sharpened by Loaded Language, Appeal to Fear: language that can make the dispute feel more urgent, personal, or adversarial than the underlying facts alone.

Follow-up questions

What new context would change how readers understand this AI National Security Risks story?
What evidence would most clearly confirm or weaken the claim that OpenAI and Anthropic briefed staffers on Thursday in two separate classified briefings?
How does this story connect AI National Security Risks with US-China AI Competition over the next few days?

open_in_new Read the original article: https://axios.com/2026/04/28/openai-anthropic-congress-cyber-briefings

analyticsAnalysis

30%

Propaganda Score

confidence: 90%

Minor concerns. Some persuasive language detected, but largely factual.

psychologyPropaganda Techniques Detected

eFinder identified 2 propaganda techniques in this article. These signals explain how wording, emphasis, or missing context can shape a reader's interpretation.

warning

Loaded Language 70% confidence

Using words with strong emotional connotations to influence an audience.

Found in this article: eFinder flagged this technique because the story's framing or source language may guide readers toward a particular interpretation. Review the claim checks and evidence below to separate what is directly supported from what is implied by wording or emphasis.

Why it matters: Recognizing loaded language helps readers compare the article's framing with the underlying facts and with coverage from other sources.

warning

Appeal to Fear 80% confidence

Building support by instilling anxiety or panic in the audience.

Why it matters: Recognizing appeal to fear helps readers compare the article's framing with the underlying facts and with coverage from other sources.

fact_checkClaims Checked

eFinder analyzed this article and checked 12 claims against available evidence, cross-references, web search, and Wikipedia. Here is what the fact-checking layer found.

check_circle Corroborated 6

info Single Source 3

schedule Pending 2

help Insufficient Evidence 1

info

Claim 1: “OpenAI and Anthropic briefed staffers on Thursday in two separate classified briefings”

SINGLE SOURCE

While the briefings are corroborated in claim 0, the specific detail about 'two separate classified briefings on Thursday' is not explicitly detailed across multiple independent sources in the provided evidence; the provided evidence for this specific claim is generic company/person info.

menu_book

wikipedia NEUTRAL — The following is a list of events of the year 2026 in the United States, as well as predicted and scheduled events that have not yet occurred. July 4, 2026, will be the 250th anniversary of the signin…
https://en.wikipedia.org/wiki/2026_in_the_United_States

menu_book

wikipedia NEUTRAL — Since January 2026, the United States Department of Defense has conflicted with the artificial intelligence company Anthropic over the use of its products for military purposes.
https://en.wikipedia.org/wiki/Anthropic–United_States_Depart…

menu_book

wikipedia NEUTRAL — Peter Brian Hegseth (born June 6, 1980) is an American government official and former television personality who has served since 2025 as the 29th United States secretary of defense. Hegseth studied p…
https://en.wikipedia.org/wiki/Pete_Hegseth

+ 3 more evidence sources

check_circle

Claim 2: “Anthropic has held off on a public release of its Mythos Preview model due to its ability to quickly find and exploit critical security flaws.”

CORROBORATED

Three independent web search results confirm Anthropic delayed the public release of the Mythos Preview model due to its ability to find and exploit security flaws.

menu_book

wikipedia NEUTRAL — The following is a list of events of the year 2026 in artificial intelligence, as well as predicted and scheduled events that have not yet occurred.
https://en.wikipedia.org/wiki/2026_in_artificial_intelligenc…

menu_book

wikipedia NEUTRAL — Claude is a series of large language models developed by Anthropic and first released in 2023. Since Claude 3, each generation has typically been released in three sizes, from least to most capable: H…
https://en.wikipedia.org/wiki/Claude_(language_model)

menu_book

wikipedia NEUTRAL — GPT-5.5 (Generative Pre-trained Transformer 5.5) is a large language model (LLM) released by OpenAI on April 23, 2026. The model is also known by its codename "Spud". OpenAI reports improvements on be…
https://en.wikipedia.org/wiki/GPT-5.5

+ 3 more evidence sources

info

Claim 3: “The briefings also touched on a recent White House memo accusing China of "industrial-scale" campaigns to distill and copy American AI models”

SINGLE SOURCE

The search results for this claim returned definitions of the color 'white' and are completely irrelevant to the claim about a White House memo on China.

travel_explore

web search NEUTRAL — White objects fully (or almost fully) reflect and scatter all the visible wavelengths of light. White on television and computer screens is created by a mixture of red, blue, and green light. The colo…
https://en.wikipedia.org/wiki/White

travel_explore

web search NEUTRAL — The meaning of WHITE is having the color of new snow or milk; specifically : of the color white. How to use white in a sentence.
https://www.merriam-webster.com/dictionary/white

travel_explore

web search NEUTRAL — WHITE definition: of the color of pure snow, of the margins of this page, etc.; reflecting nearly all the rays of sunlight or a similar light. See examples of white used in a sentence.
https://www.dictionary.com/browse/white

info

Claim 4: “OpenAI said the briefing was one of several the company held with Senate and House committees last week — in addition to a briefing with the White House.”

SINGLE SOURCE

The search results for this claim are about an 'Impact Investment Group' in Australia and are completely irrelevant to OpenAI's briefings.

travel_explore

web search NEUTRAL — About Us Impact Investment Group is a leading Australian impact investment funds manager. We want to prove that finance can be a force for good. That finance can be great for people and great for the …
https://www.impact-group.com.au/about/

travel_explore

web search NEUTRAL — Contact Us enquiries@impact-group.com.au +61 3 8534 8060 Impact Investment Group’s head office is on the traditional lands of the Boon Wurrung people, part of the Kulin nation. We acknowledge their el…
https://www.impact-group.com.au/contact-us/

travel_explore

web search NEUTRAL — Impact Investment Group is a leading Australian Impact Funds Manager, with investments in Renewable Energy, and Green Real Estate.
https://www.impact-group.com.au/

schedule

Claim 5: “Members of the committee said another briefing last week, on jailbroken AI models... gave them new urgency on regulating AI.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

check_circle

Claim 6: “House Homeland Security Chair Andrew Garbarino (R-N.Y.) has been hosting ongoing private roundtables with tech and AI executives”

CORROBORATED

Two separate web search results confirm that House Homeland Security Chair Andrew Garbarino has been hosting private roundtables with tech and AI executives.

travel_explore

web search NEUTRAL — The big picture: House Homeland Security Chair Andrew Garbarino (R-N.Y.) has been hosting ongoing private roundtables with tech and AI executives, according to the Washington Post, and has been discus…
https://www.axios.com/2026/04/28/openai-anthropic-congress-c…

travel_explore

web search NEUTRAL — Andrew Reed Garbarino is an American attorney and politician serving as the U.S. representative for New York's 2nd congressional district since 2021. A member of the Republican Party, he served as the…
https://en.wikipedia.org/wiki/Andrew_Garbarino

travel_explore

web search NEUTRAL — House Homeland Security Chair Andrew Garbarino has been holding private roundtables with tech and AI leaders, the report mentioned, adding that the committee has also held multiple hearings on generat…
https://www.aa.com.tr/en/americas/openai-anthropic-brief-us-…

schedule

Claim 7: “Members were shown ways such tools could be used to carry out a school shooting or a bombing.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

check_circle

Claim 8: “OpenAI and Anthropic briefed House Homeland Security Committee staff on their new cyber-capable AI models and their implications for cybersecurity”

CORROBORATED

Multiple web search results confirm that OpenAI and Anthropic briefed the House Homeland Security Committee on new AI models and cyber threats.

menu_book

+ 3 more evidence sources

help

Claim 9: “Rep. Jay Obernolte (R-Calif.) who introduced a bill laying out a federal framework for AI this week.”

INSUFFICIENT EVIDENCE

No evidence was found in the provided search results regarding Rep. Jay Obernolte introducing an AI framework bill in April 2026.

check_circle

Claim 10: “The committee has also held several hearings on the implications of generative AI models on national security, including nation-state cyberattacks.”

CORROBORATED

Web search results mention that the committee has held multiple hearings on generative AI's national security risks, including state-sponsored cyberattacks.

check_circle

Claim 11: “OpenAI decided on a tiered approach for releasing its GPT-5.4-Cyber model.”

CORROBORATED

Web search results confirm the release of GPT-5.4-Cyber and that it is being rolled out in a limited, iterative manner to vetted vendors and researchers (a tiered approach).

travel_explore

web search NEUTRAL — On April 15, OpenAI released GPT-5.4-Cyber. Exactly seven days after Anthropic restricted Claude Mythos to roughly 40 organizations under Project Glasswing. The timing isn’t subtle.
https://onemetrik.com/market-insights/gpt-54-cyber/

travel_explore

web search NEUTRAL — Today, we’re releasing GPT‑5.4 in ChatGPT (as GPT‑5.4 Thinking), the API, and Codex. It’s our most capable and efficient frontier model for professional work.
https://openai.com/index/introducing-gpt-5-4/

travel_explore

web search NEUTRAL — Customers in the highest tier will get access to the new cybersecurity-focused GPT-5.4-Cyber model. For now, OpenAI said GPT-5.4-Cyber will roll out in a limited and iterative manner to vetted securit…
https://www.neowin.net/news/openai-challenges-anthropics-str…

check_circle

Claim 12: “Both companies are working with federal agencies to get them access to the models.”

CORROBORATED

Multiple sources confirm both companies are providing AI access to federal agencies, including offering chatbots at $1 per year and creating models tailored for government employees.

menu_book

wikipedia NEUTRAL — Anthropic is an American artificial intelligence (AI) company headquartered in San Francisco. It has developed a range of large language models (LLMs) named Claude and focuses on AI safety. Anthropic …
https://en.wikipedia.org/wiki/Anthropic

menu_book

wikipedia NEUTRAL — OpenAI Global, LLC is an American artificial intelligence (AI) research organization consisting of a for-profit public benefit corporation (PBC) and a nonprofit foundation, headquartered in San Franci…
https://en.wikipedia.org/wiki/OpenAI

+ 3 more evidence sources

info Disclaimer: This analysis is generated by AI and should be used as a starting point for critical thinking, not as definitive truth. Claims are verified against publicly available sources. Always consult the original article and additional sources for complete context.