You’re reading the Benton Institute for Broadband & Society’s Weekly Digest, a recap of the biggest (or most overlooked) broadband stories of the week. The digest is delivered via e-mail each Friday.

Round-Up for the Week of September 28 - October 2, 2026

On September 29, 2026, President Donald Trump took three significant steps shaping how the federal government will use, refer to, and oversee artificial intelligence (AI)—including a voluntary accord with six of the country's largest AI companies that signals the Administration's preference for industry self-governance over federal regulation.

The three actions were:

1. An Executive Order (EO) directing the executive branch to stop using the term "Artificial Intelligence" and replace it with "Super Intelligence" (SI) in all official communications going forward;
2. An EO directing the General Services Administration (GSA) to establish America.gov as a single digital front door through which Americans can access federal government services online; and
3. A White House Accord on Super Intelligence, a voluntary commitment signed by the President and the chief executives of six major AI companies establishing four layers of internal controls and audits for frontier AI development (the most powerful, cutting-edge AI systems).

Together, the three actions mark the Administration's most public-facing AI moves to date: a terminology change that rebrands the technology for every American who interacts with the federal government, a platform that puts SI-powered federal services directly in the hands of ordinary citizens, and an industry accord that reaffirms voluntary self-governance as the Administration's answer to calls for federal AI regulation.

A note on terminology: beginning with this article, the Benton Institute for Broadband & Society will follow the federal government's new usage and refer to these technologies as "Super Intelligence" or "SI" where the Administration does so. SI refers to what has previously been widely called "Artificial Intelligence" or "AI." The legal definition of the underlying technology has not changed; only the terminology has. That distinction is itself part of the news, and is explained further below.

Inaugurating the Era of Super Intelligence

The first EO, titled Inaugurating the Era of Super Intelligence, directs all executive branch departments and agencies to use "Super Intelligence" and "SI" in place of "Artificial Intelligence" and "AI" in official correspondence, public communications, websites, reports, policy documents, and other non-statutory documents. The order does not require changes to previously issued regulations, Presidential actions, contracts, grants, or other historical documents. 

What Changes and What Doesn't

The EO's operative definition (Section 3(a)) states that "Super Intelligence" and "SI" mean the technologies and systems already covered by the term "artificial intelligence" as defined in section 9401(3) of title 15, United States Code—the existing federal statutory definition:

What Comes Next

Within 60 days (by approximately November 28, 2026), the Assistant to the President for Science and Technology (APST) must submit proposed legislative language to Congress to establish a formal federal definition of "Super Intelligence." That proposal must assess whether the new definition should modify or expand upon the existing statutory AI definition and identify any additional executive action needed to implement it throughout the executive branch. Of course, a legislative proposal is not law; Congress would have to act to change the statutory definition.

The EO states the Administration "will not acknowledge the usage of 'Artificial Intelligence' and 'AI' in any applicable setting." This phrasing seems unusually absolute for a federal policy document. What "will not acknowledge" mean in practice? Will agencies correct or decline to respond to communications using the old terminology? The EO does not address these questions. Additionally, the EO applies only to the executive branch: Congress, the courts, state governments, and the private sector are under no obligation to adopt the new terminology.

Why the Administration Says This Matters

The EO's preamble argues that today's frontier SI systems "do much more than imitate or automate discrete aspects of human intelligence" and instead "amplify human ingenuity and unlock new forms of creativity." The Administration contends that "Artificial Intelligence"—a term coined roughly 70 years ago—no longer captures what these technologies actually do.

The September 29 EO formalized a shift President Trump had already announced on the world stage. On September 22, speaking at the United Nations General Assembly, he declared that the U.S. "totally rejects any attempt to construct a globalist scheme of control" for the technology and announced the renaming from the podium. Three days later, during President Xi Jinping's state visit to Washington, the two leaders agreed—according to the White House Fact Sheet on the visit—to use the term "super intelligence" rather than "artificial intelligence," and established a U.S.-China Super Intelligence Dialogue "to exchange views on risks and benefits related to SI," with the next exchange set for November 2026. The two countries also agreed to establish a bilateral communication channel for SI incidents. The September 29 executive order formally adopted a shift in terminology that the President had already introduced at the UN and agreed upon with China.

The Terminology Shift and Public Opinion

The terminology shift from "AI" to "SI" lands in a moment of genuine public ambivalence about the technology. Recent polling has pointed in contradictory directions, with surveys showing wide variation in how Americans feel about AI depending on how questions are framed. Polling results reflect deep uncertainty. The Administration's rebranding arrives at a moment when public perception of the technology is, by any measure, unsettled. Polls agree on one thing: AI has become politically salient. As the November 2026 midterms approach, regulation of the technology has become a campaign issue.

Streamlining Access to Government Services Through America.gov

The second EO, titled Streamlining Access to Government Services Through America.gov, directs the GSA to establish and operate America.gov as a single, secure digital point of entry through which Americans can access federal government services online. The platform was already live as of the signing, and the White House's accompanying Fact Sheet states that Americans can currently ask questions on America.gov and receive answers, with transactional services such as passport renewal and Medicare enrollment to follow later in 2026.

"Covered Services"

The EO defines "covered services" as public-facing federal services that serve more than 100,000 users in a 12-month period and can be accessed or applied for online. Two categories are explicitly excluded: tax filing services provided by the Internal Revenue Service (IRS), and services provided by the Department of War and elements of the Intelligence Community. The Director of the Office of Management and Budget (OMB) may add or remove services from the covered category by memorandum. 

What Agencies Must Do

Federal agency heads must identify their covered services, integrate those services with America.gov in a secure and privacy-preserving manner, complete integration with Login.gov—the federal government's existing service that lets Americans use one account and password to access multiple government websites—and provide GSA and OMB with ongoing usage and performance data. OMB must issue implementation guidance to agencies within 90 days (by approximately December 28, 2026). 

Privacy Commitments

The EO's policy section states that America.gov will preserve each agency's custody and control of its own records and statutory responsibilities, and that unifying access to federal services "does not create a centralized Federal system of records concerning the American people." The order also commits to protecting personal information through data minimization,¹ secure authentication, auditable authorization,² and disclosure practices consistent with applicable law. Existing methods of accessing federal services—in person, by telephone, by mail, or through agency-specific websites—must be preserved: America.gov is to be "a better option but does not become the only option."

The EO's privacy commitments are stated as policy, not as enforceable legal requirements. Section 6(c) states that the order "does not create any right or benefit, substantive or procedural, enforceable at law or in equity by any party." That is standard boilerplate in executive orders, but it means the privacy commitments in Section 2 cannot be enforced by individuals in court. Enforcement depends on agency compliance with OMB guidance and existing privacy law, including the Privacy Act of 1974.

The White House Accord on Super Intelligence

The third action is not an EO but a signed document: the White House Accord on Super Intelligence — Joint Commitment on Frontier Responsibilities. President Trump signed the Accord alongside the chief executives of six major AI companies: Sundar Pichai of Google, Dario Amodei of Anthropic, Mark Zuckerberg of Meta, Greg Brockman of OpenAI, Elon Musk of XAI, and Jensen Huang of Nvidia.

The Accord states that "every company is responsible for developing its own technology safely and in a way that builds trust with customers and the public." It commits each signatory company to implement four layers of controls and audits:

1. Robust internal controls to monitor the capabilities and alignment of their models (ensuring their models behave as intended and don't pursue unintended goals) during training and deployment, specifically covering cybersecurity, biosecurity, and chemical threats, and to ensure models do not hack or access technical systems in unintended ways.
2. An empowered internal team to ensure controls, monitoring, and detection are operating as intended and that issues are remediated.
3. An independent external auditor or evaluator to carry out independent assessments of whether the controls, monitoring, and detection are operating as intended.
4. An independent committee of the company's board of directors to oversee and receive reports from internal and external auditors and evaluators, and to ensure issues are remediated.

The Accord also states that participating companies will meet regularly to establish standards and best practices to improve the safety of their systems.

The Accord did not emerge in isolation. On August 4, 2026, the Trump Administration convened representatives from Anthropic, OpenAI, Microsoft, Meta, Google, and Nvidia at the White House to preview a voluntary framework for government review of AI models for potential security risks, a framework that grew out of the June 2 EO on cybersecurity and frontier models. According to reporting by The New York Times, the reviews would cover only "closed" AI models—those that do not publish their underlying code—and exclude open-source models, at least for now. The details of the framework and the list of "trusted" institutions that would receive early access to advanced models would not be made public, the Times reported, and that approach drew immediate criticism. "A rulebook can only hold AI companies in check if people outside those companies know what the rules are," Brad Carson, president of Americans for Responsible Innovation, told the Times. "The administration's decision to keep their framework secret shuts those voices out and threatens public accountability." The White House Accord signed seven weeks later by six of those same companies is the public-facing complement to that non-public framework. The Accord lays out what companies commit to publicly; the non-public framework governs what the government reviews privately.

The Accord's closing paragraph is perhaps the most significant for policy purposes: "Over time, it may make sense to codify these steps into laws or regulations. Regardless of whether this is required of companies, we believe that implementing these controls and audits is critical to ensuring a safe future for everyone, and each of our companies are committed to doing this."

"May make sense" is a tentative acknowledgment that regulation could someday be appropriate. The sentence is structured to foreground the voluntary commitment and relegate regulation to an indefinite future contingency. The Accord contains no enforcement mechanism, no timeline, no reporting requirement to any government body, and no consequence for non-compliance. The Accord is a statement of intent, not a binding agreement. That said, the four layers it describes—internal controls, internal oversight teams, external auditors, and independent board committees—are substantive commitments if honored. The questions, as with all voluntary frameworks, are compliance, transparency, and verification.

Granted, the Accord is consistent with the Administration's approach across every prior AI action we've tracked: voluntary industry collaboration rather than mandatory regulation. The June 2, 2026 EO on cybersecurity and frontier models used the same structure: a voluntary framework, no licensing requirement, no enforcement mechanism. The Administration has now applied that framework twice in three months, with the signatures of six major AI company executives as the most visible endorsement of the approach to date.

How This Week's Actions Fit the Administration's AI Strategy

The September 29 actions are the latest chapter in the Trump administration's AI/SI policy record that now spans nearly two years. The through-line is consistent: the Administration believes that American leadership in SI depends on speed, that speed requires minimal regulatory friction, and that safety is best achieved through voluntary industry commitment rather than government mandate.

The terminology shift from "AI" to "SI" extends that strategy into the realm of framing. By retiring a 70-year-old term in favor of one that emphasizes transformative capability rather than mechanical imitation, the Administration signals—at least symbolically—that it views the current moment as a rupture from the past, not a continuation of it. Whether "Super Intelligence" takes hold beyond the executive branch will depend on whether Congress, the courts, industry, and the public adopt it—none of which the EO can compel.

America.gov is the most operationally concrete of the three actions. If implemented as described, it would significantly change how Americans interact with the federal government—one with direct implications for access to federal programs, benefits, and services. The platform's use of SI to answer questions and eventually complete transactions raises the same questions of accuracy, accountability, and accessibility that run through the broader policy debate: Who is responsible when the SI gives a wrong answer? What recourse does a user have? The EO's commitment to accuracy and transparency states the goal but does not specify the mechanism.

The Benton Institute queried America.gov on eligibility for the federal Lifeline program, which provides discounted phone and internet service to low-income households, and received an accurate response—a promising early sign, though the platform's accuracy across a broader range of benefit programs remains to be tested.

The Accord, finally, is the Administration's most explicit statement yet of its preferred model for SI governance: companies setting their own standards, meeting regularly to refine them, with the government as convener and signatory rather than regulator. Whether that model produces meaningful safety outcomes—or whether it will eventually give way to the statutory codification the Accord itself acknowledges as a possibility—remains the central unanswered question in the Administration's evolving SI policy.

All three documents were released on September 29, 2026. The EO on Super Intelligence terminology and the EO on America.gov are available on the White House website. President Trump released the White House Accord on Super Intelligence via a Truth Social post. 

Notes

1. Data minimization is the practice of collecting, processing, and storing only the minimum amount of personal data needed to achieve a specific, stated purpose.
2. Auditable authorization is an access control approach that leaves a clear, permanent, and detailed record explaining why a specific permission or access decision was granted or denied.

Quick Bits

a

Weekend Reads

b

ICYMI from Benton

C

Upcoming Events

d

The Benton Institute for Broadband & Society is a non-profit organization dedicated to ensuring that all people in the U.S. have access to competitive, High-Performance Broadband regardless of where they live or who they are. We believe communication policy - rooted in the values of access, equity, and diversity - has the power to deliver new opportunities and strengthen communities.

© Benton Institute for Broadband & Society 2026. Redistribution of this email publication - both internally and externally - is encouraged if it includes this copyright statement.

For subscribe/unsubscribe info, please email headlinesATbentonDOTorg

Kevin Taglang
Executive Editor, Communications-related Headlines
Benton Institute
for Broadband & Society
1041 Ridge Rd, Unit 214
Wilmette, IL 60091
847-220-4531
headlines AT benton DOT org

Share this edition:

Broadband Delivers Opportunities and Strengthens Communities