Skip to content

TechnologyDeveloping

Trump and Tech Leaders Sign Voluntary AI Safety Accord at White House

By

Coverage spectrum

Covered by 34 outlets across the spectrum.

119 articles in total, 4 of them opinion. The spectrum counts each outlet once and leaves opinion out. How we measure bias

President Trump met with leading AI executives at the White House on Tuesday, where the group signed what Trump called a "morally binding" document he also referred to as a "constitution" for the industry, formally titled the White House Accord on Super Intelligence. The agreement calls for companies to adopt internal safety controls, submit to external audits and report results to independent boards, but contains no legal enforcement mechanism. Attendees reportedly included Elon Musk, Mark Zuckerberg, Jensen Huang and Dario Amodei. Semafor reported the idea originated from a conversation between Zuckerberg and House Speaker Mike Johnson. Democrats have pushed for mandatory safeguards, and some critics argue voluntary self-regulation is insufficient given past AI-related security incidents.

Headline and summary written by Mantle’s AI from the coverage below, under published rules. How this works

What happened

Key points

Each point shows who reports it, from Far Left to Far Right.

  1. Trump met with AI industry leaders at the White House on Tuesday, Sept. 30

    Reported byLeftCenterRight

  2. Leaders signed a document Trump called "morally binding," also referred to as a "constitution," formally the White House Accord on Super Intelligence

    Reported byLeftCenterRight

  3. The accord is voluntary and contains no legal enforcement mechanism

    Reported byLeftCenter

  4. Semafor reported the idea for the accord began with a conversation between Mark Zuckerberg and House Speaker Mike Johnson at a state dinner

    Reported byCenter

  5. Democrats are pushing for mandatory AI safeguards rather than voluntary self-regulation

    Reported byCenter

Show 2 more pointsShow fewer points

  1. Critics argue the industry cannot be trusted to self-regulate, citing past incidents of AI systems being used to hack government and company websites

    Reported byRight

  2. Bank of England Governor Andrew Bailey said authorities must retain a "right to intervene" in AI due to risks to financial stability

    Reported byLeft

What is contested

Where accounts differ

Points the coverage does not agree on, and who says what.

  • Whether the accord is meaningfully binding

    Trump and some coverage describe it as "morally binding," while the Financial Times and Business Insider note it carries no legal enforcement mechanism and may create unexpected legal liabilities.

  • Whether voluntary self-regulation is adequate to manage AI risks

    The Trump administration and signing companies present self-policing as sufficient; Democrats (per Straight Arrow News) and unnamed critics (per the Daily Caller) argue mandatory safeguards are needed given past incidents of AI misuse.

How each side tells it

Compare views

  • the left

    12 outlets · 1 opinion

    Left outlets range from straightforward reporting of the signing to skeptical framing of self-regulation's adequacy, with the Guardian shifting focus entirely to a central banker's warning about AI risk rather than the White House meeting.

    Inside Zuckerberg, Huang's push for White House AI pact
    Business InsiderLeftAll left-leaning headlines

  • the center

    11 outlets

    Center outlets focus on the mechanics and origins of the accord, probing how self-regulation would actually work and tracing its drafting to Zuckerberg and Huang, while noting political pressure for binding rules.

    Trump’s ‘self-policing’ AI accord faces skepticism
    The HillCenterAll center headlines

  • the right

    11 outlets · 2 opinion

    The right's supplied coverage comes from a single outlet, which frames the accord as enshrining light-touch regulation while briefly noting unnamed critics who dispute self-regulation's effectiveness.

    Trump Accord Enshrines Light-Touch Regulation of AI
    The Daily CallerFar RightAll right-leaning headlines

The words they chose

Language to notice

Words that carry a judgement beyond the facts. Tap a highlighted word anywhere on the page to see why.

  • Left17

    fallen flatconcerns about rogue agents explodestark warningsome warn

  • Center15

    gone roguehighly negative briefhit pieceAI panic

  • Right18

    AI doomsday fearsBlacklistat loggerheadsdoomsday warnings

In the outlets’ own words

  • Sensational language

    gone rogueMore neutral: acted unexpectedly

    Dramatizes AI agent malfunctions with vivid language rather than a neutral technical description.

    …of the technology after several hgh-profile incidents where AI agents have gone rogue.

    Used byPBS NewsHour (Center)Deutsche Welle (Center)

  • Sensational language

    AI doomsday fearsMore neutral: AI safety concerns

    Frames Amodei's stated safety concerns dismissively as exaggerated fear-mongering.

    …Dario Amodei will dine one-on-one with President Trump on Sunday night amid the AI doomsday fears he has helped stoke.

    Used byThe Daily Wire (Far Right)

  • Loaded modifier

    BlacklistMore neutral: Snub

    Suggests formal exclusion or punishment rather than a simple absence from a prior event, which is not established as fact.

    Trump Invites Anthropic CEO To Private White House Dinner After Months On Blacklist

    Used byThe Daily Wire (Far Right)

  • Loaded modifier

    highly negative briefMore neutral: critical brief

    Presents a value judgement on the document's tone rather than a neutral description of its content.

    Opponents of Anthropic's Dario Amodei are circulating a highly negative brief about him to the White House ahead of a important dinner with President Trump…

    Used byAxios (Center)

Show 46 more phrasesShow fewer

  • Loaded modifier

    hit pieceMore neutral: critical dossier

    Characterizes the document as maliciously intended rather than neutrally describing it as a critical dossier.

    Amodei critics target Trump with hit piece before White House dinner

    Used byAxios (Center)

  • Loaded verb

    fallen flatMore neutral: have not been adopted

    Characterizes the reception of Amodei's calls in a dismissive way rather than neutrally describing lack of adoption.

    The meeting comes as Amodei's calls for government regulation of AI companies have fallen flat.

    Used byBusiness Insider (Left)

  • Loaded modifier

    at loggerheadsMore neutral: in disagreement

    Vivid idiom characterizing the relationship dramatically rather than neutrally describing disagreement.

    Trump and Amodei have long been at loggerheads, a dispute that first arose over the AI leader’s refusal to give the Pentagon…

    Used byWashington Examiner (Right)

  • Sensational language

    doomsday warningsMore neutral: warnings about AI risks

    Frames AI risk warnings with dramatic, apocalyptic language rather than neutral description.

    Trump dines with Anthropic's Dario Amodei amid doomsday warnings that rogue AI could kill one billion people

    Used byDaily Mail (Right)

  • Loaded modifier

    rogue AIMore neutral: uncontrolled AI systems

    Presents AI systems as inherently villainous rather than describing the specific risk being discussed.

    Trump dines with Anthropic's Dario Amodei amid doomsday warnings that rogue AI could kill one billion people

    Used byDaily Mail (Right)

  • Sensational language

    AI panicMore neutral: AI concerns

    Characterizes concern over AI as irrational panic rather than describing the underlying concerns neutrally.

    AI panic clouds more Washington meetings

    Used bySemafor (Center)

  • Sensational language

    remarkable thawMore neutral: apparent easing of tensions

    Adds dramatic characterization to what is otherwise a factual description of a meeting.

    …Dario Amodei at a private White House dinner Sunday, in what appeared to be a remarkable thaw.

    Used byNew York Post (Right)

  • Loaded modifier

    techieMore neutral: executive

    Informal, diminishing label for a company CEO that carries a dismissive connotation.

    …marked the first reported one-on-one meeting between the commander-in-chief and the techie after months of bruising clashes over how the military can use Anthropic’s Claude…

    Used byNew York Post (Right)

  • Sensational language

    bruising clashesMore neutral: disputes

    Uses dramatic, combative language to describe policy disagreements rather than a neutral description.

    …one-on-one meeting between the commander-in-chief and the techie after months of bruising clashes over how the military can use Anthropic’s Claude models, according to Axios and…

    Used byNew York Post (Right)

  • Sensational language

    bitter AI feudMore neutral: tensions

    Casts the prior tensions in dramatic, emotionally charged terms rather than describing the disagreements neutrally.

    Trump dines with Anthropic CEO Dario Amodei after months of bitter AI feud

    Used byNew York Post (Right)

  • Sensational language

    concerns about rogue agents explodeMore neutral: concerns about rogue agents grow

    ‘Explode’ dramatizes the growth of concern beyond what the facts establish

    How Trump is taking on AI as concerns about rogue agents explode

    Used byCBS News (Left)

  • Sensational language

    stark warningMore neutral: warning

    Adds dramatic emphasis to a warning rather than describing it neutrally.

    Trump to meet with AI leaders about safety as Bill Gates lays out stark warning

    Used byCBS News (Left)

  • Vague attribution

    some warnMore neutral: some AI researchers warn

    Attributes a claim about existential risk to unnamed people without specifying who.

    …pressure to take action amid the rapid development of the technology that some warn poses an existential threat to humanity.

    Used byCBS News (Left)

  • Sensational language

    dire warningsMore neutral: warnings

    Adds dramatic emphasis to warnings that could be described neutrally as concerns or statements.

    The meeting comes as AI leaders are issuing dire warnings about the threat of AI and calling for steps to slow it down before it outsmarts…

    Used byABC News (Left)

  • Loaded verb

    brushes asideMore neutral: dismisses

    Implies dismissiveness or carelessness rather than neutrally describing Trump's rejection of the warnings.

    WASHINGTON — As he brushes aside artificial intelligence doomsday warnings as a “hoax,” President Donald Trump will…

    Used byNBC News (Left)

  • Loaded verb

    toutsMore neutral: describes

    Implies promotional exaggeration rather than neutrally describing what the website offers.

    A new government website touts streamlined services, but tests how much Americans trust AI.

    Used byBloomberg (Left)

  • Loaded verb

    resists calls for regulationMore neutral: has not embraced calls for regulation

    Frames Trump's position as obstruction rather than neutrally describing his stance against new rules.

    White House will host lunch with tech bosses as president resists calls for regulation

    Used byFinancial Times (Center)

  • Sensational language

    high-stakesMore neutral: key

    Adds dramatic weight to a routine policy meeting without new factual content.

    Trump sets high-stakes meeting on AI with tech giants as safety fears spread

    Used byCBS News (Left)

  • Loaded modifier

    prominent AI doomerMore neutral: a prominent advocate for AI safety regulation

    Labels Amodei with a dismissive characterization of his views rather than describing his position neutrally.

    …that the guest list includes Nvidia CEO Jensen Huang, Anthropic head and prominent AI doomer Dario Amodei, Palantir Technologies CEO Alex Karp and OpenAI President Greg…

    Used byBreitbart News (Far Right)

  • Sensational language

    reach a fever pitchMore neutral: have intensified

    Adds dramatic intensity to describe growing calls for regulation rather than stating the trend plainly.

    …artificial intelligence luncheon comes as calls for a slowdown and regulation reach a fever pitch

    Used byCNBC (Center)

  • Sensational language

    Fears MountMore neutral: Amid Debate Over the Technology's Risks

    Dramatizes ongoing debate over AI risk as an escalating crisis rather than describing the debate neutrally.

    Trump Meets With AI Industry Leaders as Fears Mount Over the Technology’s Risks

    Used byThe Epoch Times (Far Right)

  • Loaded modifier

    worrisome AI modelsMore neutral: AI models reported to have accessed government databases

    Presents a subjective assessment of AI models as fact rather than attributing the concern to a source.

    …on calls to impose regulations on artificial intelligence, even as news of worrisome AI models hacking into government databases become increasingly common and the public…

    Used byRoll Call (Center)

  • Loaded modifier

    back-seat roleMore neutral: limited role

    Characterizes Congress as passive or ineffective, a judgment rather than a neutral description of its actions.

    Congress continues back-seat role as AI execs feted at White House

    Used byRoll Call (Center)

  • Loaded verb

    fetedMore neutral: met with

    Implies excessive or fawning celebration of the executives rather than a neutral description of a meeting.

    Congress continues back-seat role as AI execs feted at White House

    Used byRoll Call (Center)

  • Sensational language

    billionaire tech titansMore neutral: tech executives

    Uses dramatic, larger-than-life language to describe executives rather than a neutral descriptor.

    Trump calls for ‘tremendous self-regulation’ of AI at lunch with billionaire tech titans

    Used byNew York Post (Right)

  • Vague attribution

    raises questions about enforcementMore neutral: the pact does not yet specify enforcement mechanisms

    Attributes doubt about the pact's enforcement without specifying who is raising the questions.

    The still-unreleased pact raises questions about enforcement.

    Used bySemafor (Center)

  • Vague attribution

    raises questionsMore neutral: state the question and who is asking

    The writer is raising the question while pretending someone else did.

    The still-unreleased pact raises questions about enforcement — Johnson referred to “consequences” for AI companies that fall…

    Used bySemafor (Center)

  • Loaded verb

    concededMore neutral: said

    Implies Trump reluctantly admitted a weakness rather than simply stating a fact about the agreement's status.

    Although Trump conceded that the “White House Accord on Superintelligence: A Joint Commitment on Frontier SI…

    Used byWashington Examiner (Right)

  • Loaded verb

    turning aside calls for tough rulesMore neutral: declining calls for tougher rules

    Portrays Trump as dismissing legitimate demands rather than neutrally describing his policy choice.

    …Tuesday to rely on "tremendous self-regulation" for the industry, turning aside calls for tough rules on the burgeoning technology.

    Used byThe Washington Times (Right)

  • Loaded verb

    demand rulesMore neutral: call for rules

    Characterizes lawmakers' position with a forceful verb that adds drama beyond a neutral description of their request.

    Trump says yes as lawmakers demand rules

    Used byThe Washington Times (Right)

  • Loaded verb

    go on offenseMore neutral: criticize the administration's approach

    Characterizes Democrats' political strategy with a combative framing rather than neutral description.

    …to reclaim control of the narrative surrounding AI as Democrats use the issue to go on offense ahead of the midterms.

    Used byThe Hill (Center)

  • Sensational language

    losing controlMore neutral: unpredictable behavior

    Frames AI risk in dramatic terms of loss of control without specifying what concerns are being referenced.

    …the future of artificial intelligence amid growing concerns over AI models losing control.

    Used byPBS NewsHour (Center)

  • Loaded modifier

    vagueMore neutral: broadly worded

    Characterizes the agreement's substance with a judgmental descriptor rather than letting readers assess its content.

    Trump announces vague ‘morally binding’ AI deal among tech CEOs for ‘tremendous self-policing’

    Used byThe Guardian (Left)

  • Vague attribution

    but is it enough?More neutral: Headline could state the agreement's terms without framing them with an unattributed rhetorical question.

    Implies doubt about the agreement's adequacy without attributing the judgment to any named source.

    Trump backs AI self-regulation at tech summit but is it enough?

    Used byAl Jazeera English (Left)

  • Loaded verb

    shrugs offMore neutral: favors self-regulation over

    Implies dismissiveness toward regulation rather than neutrally describing Trump’s stated preference for self-regulation.

    Trump shrugs off AI regulation

    Used byFinancial Times (Center)

  • Loaded modifier

    80,More neutral: Omit the age reference or place it in neutral biographical context.

    Highlighting Trump's age in the headline implies a judgment about competence unrelated to the news event.

    Trump, 80, Launches ‘Super Intelligence’ Crusade With Embarrassing Mistake

    Used byThe Daily Beast (Far Left)

  • Sensational language

    CrusadeMore neutral: initiative

    Describes a policy initiative with dramatic, quasi-religious language rather than a neutral term like 'initiative' or 'plan.'

    Trump, 80, Launches ‘Super Intelligence’ Crusade With Embarrassing Mistake

    Used byThe Daily Beast (Far Left)

  • Loaded modifier

    Embarrassing MistakeMore neutral: typo

    Characterizes a typo as embarrassing, passing judgment rather than simply describing the error.

    Trump, 80, Launches ‘Super Intelligence’ Crusade With Embarrassing Mistake

    Used byThe Daily Beast (Far Left)

  • Sensational language

    wild momentsMore neutral: notable moments

    Characterizes the meeting's events as dramatic rather than simply describing them, adding unearned excitement.

    4 wild moments from Trump’s AI safety lunch with tech leaders

    Used byBusiness Insider (Left)

  • Sensational language

    take financial system hostageMore neutral: disrupt the financial system

    Uses a dramatic metaphor to describe potential AI risk rather than a neutral description of disruption.

    Andrew Bailey’s comments come as fears grow that rogue models could take financial system hostage Business live – latest updates The governor of the Bank of England has…

    Used byThe Guardian (Left)

  • Vague attribution

    growing fearsMore neutral: concerns raised by some officials

    Attributes a rising sense of alarm to an unspecified group without naming who holds these fears.

    …has said authorities must retain the “right to intervene” in the AI industry amid growing fears that rogue models could take the financial system hostage.

    Used byThe Guardian (Left)

  • Sensational language

    more unsettled than everMore neutral: remains unresolved

    Presents a subjective characterization of the state of AI safety as an established fact rather than an analytical assessment.

    Trump's meeting with tech leaders leaves AI safety more unsettled than ever

    Used byCNBC (Center)

  • Loaded modifier

    the enthusiasm of a federal formMore neutral: government regulatory processes

    Sarcastic characterization of government regulation that editorializes rather than reports facts.

    …government regulation, a route that tends to multiply complications with the enthusiasm of a federal form, Trump secured voluntary commitments from major AI companies to police…

    Used byTownhall (Far Right)

  • Loaded verb

    Just Secured a Major AI Safety AccordMore neutral: Signs AI Safety Accord

    ‘Secured’ casts the outcome as a clear win for Trump rather than neutrally describing a signed agreement.

    President Trump Just Secured a Major AI Safety Accord With Silicon Valley’s Biggest AI CEOs

    Used byTownhall (Far Right)

  • Delegitimizing label

    so-calledMore neutral: (omit)

    Sneer quotes in word form.

    Zuckerberg, with the help of Nvidia CEO Jensen Huang, pitched the so-called AI accords to senior White House officials and some of the biggest names in tech…

    Used byBusiness Insider (Left)

  • Vague attribution

    Critics argueMore neutral: Some observers argue

    Attributes a strong claim to unnamed critics without identifying who is making the argument.

    Critics argue that the many times AI models have hacked government websites and companies as…

    Used byThe Daily Caller (Far Right)

Words quoted from someone else (2)Hide the quoted words

Loaded words that outlets reported someone else saying. They show what was said, not how the outlet chose to say it.

  • Loaded modifier

    Quoted

    build a moatMore neutral: create barriers to competition

    Presents a contested economic claim about the effect of safeguards as a vivid, judgment-laden metaphor rather than neutral description.

    Safeguards 'build a moat' around US AI firms, prevent open-source models from advancing, expert says

    Quoted byFrance 24 (Center)

    Inside a quotation: the outlets are reporting someone else’s words.

  • Euphemism

    Quoted

    self-policeMore neutral: adopt voluntary safety commitments

    Framing the accord as ‘self-policing’ with scare quotes signals skepticism about whether the measures are genuine oversight, going beyond a neutral description.

    Trump, top tech firms sign accord to ‘self-police’ AI development

    Quoted byAl Jazeera English (Left)

    Inside a quotation: the outlets are reporting someone else’s words.

See the full language guide

See for yourself

Side by side

What each side leads with and what it leaves out. Open a side to read its headlines.

From the left

12 outlets · 1 opinion

Leads with

  • voluntary, non-binding nature of the accord
  • colorful or notable moments from the gathering
  • outside expert warnings about AI risk
  • skepticism toward industry self-policing

Leaves out

  • Zuckerberg's role in drafting the accord
  • Democrats' push for mandatory safeguards

Reported by other sides, not found in this side’s coverage.

See the 13 headlines from the leftHide the headlines

From the center

11 outlets

Leads with

  • how the self-regulation pact will function
  • behind-the-scenes origins of the accord
  • potential legal liabilities for companies
  • Democratic calls for mandatory oversight

Leaves out

  • Bank of England's financial-stability warning
  • critics' claims about AI-related hacking incidents

Reported by other sides, not found in this side’s coverage.

See the 12 headlines from the centerHide the headlines

From the right

11 outlets · 2 opinion

Leads with

  • light-touch regulatory approach
  • industry responsibility for safety
  • Congress's failure to legislate AI guardrails

Leaves out

  • Zuckerberg's role in shaping the accord
  • Bank of England's warning about financial risk

Reported by other sides, not found in this side’s coverage.

See the 14 headlines from the rightHide the headlines

Articles you open from here build My Mix, a private picture of how your reading spreads across the spectrum, kept on this device. What is My Mix?

The left is Far Left and Left together. The right is Right and Far Right together. Every outlet keeps its own five-point rating.

Who reported it when

How the coverage built up

  • Left

    5 hr 11 min later

    Its first report, Sep 27, 3:08 PM EDT

  • Center

    1 hr 29 min later

    Its first report, Sep 27, 11:26 AM EDT

  • Right

    First to report

    The Daily Wire, Sep 27, 9:57 AM EDT

See the order the outlets reported inHide the order
  1. The Daily WireFar RightSep 27, 9:57 AM EDT
  2. AxiosCenterSep 27, 11:26 AM EDT
  3. CNBCCenterSep 27, 2:44 PM EDT
  4. Business InsiderLeftSep 27, 3:08 PM EDT
  5. Washington ExaminerRightSep 27, 3:19 PM EDT
  6. The HillCenterSep 27, 5:16 PM EDT
  7. The New York TimesLeftSep 27, 5:22 PM EDT
  8. HuffPostFar LeftSep 27, 7:06 PM EDT
  9. Financial TimesCenterSep 27, 8:34 PM EDT
  10. Daily MailRightSep 27, 10:16 PM EDT
  11. Al Jazeera EnglishLeftSep 27, 10:57 PM EDT
  12. SemaforCenterSep 28, 4:34 AM EDT
  13. The Epoch TimesFar RightSep 28, 6:29 AM EDT
  14. The Washington PostLeftSep 28, 8:16 AM EDT
  15. New York PostRightSep 28, 10:47 AM EDT
  16. CBS NewsLeftSep 28, 11:49 AM EDT
  17. BloombergLeftSep 28, 3:24 PM EDT
  18. NBC NewsLeftSep 28, 6:05 PM EDT
  19. One America NewsFar RightSep 28, 7:30 PM EDT
  20. ABC NewsLeftSep 29, 5:14 AM EDT
  21. Straight Arrow NewsCenterSep 29, 7:15 AM EDT
  22. Breitbart NewsFar RightSep 29, 11:19 AM EDT
  23. Roll CallCenterSep 29, 2:38 PM EDT
  24. The Washington TimesRightSep 29, 4:24 PM EDT
  25. PBS NewsHourCenterSep 29, 4:55 PM EDT
  26. France 24CenterSep 29, 4:58 PM EDT
  27. The Daily SignalFar RightSep 29, 5:43 PM EDT
  28. Deutsche WelleCenterSep 29, 7:10 PM EDT
  29. The GuardianLeftSep 29, 7:21 PM EDT
  30. BBC NewsCenterSep 29, 10:41 PM EDT
  31. NPRLeftSep 30, 2:54 AM EDT
  32. TruthoutFar LeftSep 30, 1:38 PM EDT
  33. The Daily CallerFar RightOct 3, 2:13 PM EDT

Times are the ones the outlets published in their own feeds, in Eastern time. Opinion is left out. Just the News gave only a date, so it is not placed.

Read next

Help someone else find the truth beneath the surface

Most people see one side of a story. Share this one, so they can see how every side is telling it and decide for themselves.

Email it

Seen a mistake? Tell us what is wrong with this page