Editor BGF

In complex engineering, proposing an improvement can be easier than proving that it works. When a single evaluation takes hours, testing becomes a limit on discovery itself.

DuoDiscover, introduced in a research preprint dated October 12, addresses that constraint. Its coding agents search for improvements while developing faster, task-specific measures of progress. These provisional measures are repeatedly checked against the real objective, helping prevent agents from optimizing a shortcut that no longer reflects genuine performance.

The researchers report gains of 1.92× for vLLM and 1.38× for SQLite over the base code. In a CPU microarchitecture task, where a single evaluation can take up to 28 hours, DuoDiscover surpassed both the best human design and an AlphaEvolve-based system, achieving 1.65 times that system’s gain in 16.7 days, compared with more than 40 days.

The prospect is consequential: AI agents may help engineers explore more alternatives within the time available, including changes that would otherwise be too expensive to investigate.

From BGF’s perspective, this is a promising direction for Human–AI Co-Creation in technology development. Human engineers define worthwhile objectives, assess trade-offs, and retain responsibility for validation. AI expands the range of possibilities they can examine.

For AIWS Constitutional Silicon, the opportunity is to combine such engineering capability with clear human authority and traceable verification.

The next test is whether these gains hold across broader workloads, and whether engineers understand why they work. As Beacon Papers No. 16 argues, a discovery becomes humanity’s knowledge only when people have a path to verify, interpret, and build upon it.

On October 6, OpenAI posted 722 mathematical manuscripts produced by an unreleased internal model, which the company says represent progress on 372 open problems. Only 162 came with main results checked in the Lean proof assistant. Within a day, three were withdrawn after a sign error broke one argument and two that depended on it. OpenAI acknowledged that many of the results are not yet understood by its own mathematicians.

A month earlier, on September 4, Anthropic reported that Claude, working largely autonomously for 11 days by the company’s account, had formalized Fermat’s Last Theorem in about 13 million lines of Lean. The code was released for inspection; it is machine-checkable, not a proof a human can read.

Two releases, two answers to the same new fact: machines can now produce mathematics faster than humans can referee it. One released claims ahead of verification; the other released verification that no person can read through.

The deeper question concerns humanity, not the machines. A result proven by a machine becomes humanity’s knowledge only when people can understand it, explain it, and build on it. This is the principle of Beacon Papers No. 15, The Human Advancement Test: better output does not, by itself, demonstrate human advancement. As discovery accelerates, the measure of progress is not how many theorems machines produce, but how much human understanding grows with them.

→ Read: Beacon Papers No. 15 — The Human Advancement Test

 

Work can be delegated. Answerability cannot.

At DevDay on September 29, 2026, OpenAI introduced agents that can take on ongoing responsibilities. It also opened ChatGPT as a shared surface where people and agents work together, reaching some 1.2 billion users each week.

The shift is not in capability but in role. Until now, AI performed tasks: a person asked, and the system answered. Now AI begins to hold responsibilities that continue after the person has turned away. This changes the question humanity must ask.

A responsibility that runs over time must be defined before it begins: who granted it, what it covers, and when it ends. This is the constitutional mandate set out in Beacon Papers No. 11, No Power Without a Mandate. Every consequential action taken under it must also lead to a human being who can be called to account. That is the principle of Beacon Papers No. 10, No AI Agent Without Identity: no consequential AI action without a human answerer.

As AI moves from answering requests to carrying responsibilities, the measure of progress is whether human answerability keeps pace.

Read OpenAI’s DevDay 2026 recap →

Claude now leads a quarter of Anthropic’s model R&D. Self-improvement is not self-authorization.

Anthropic disclosed on September 17 that, as of August, about 90 percent of its model research and development tasks involved collaboration with Claude — the model doing large portions of the work under close human direction. Claude now leads 26 percent of the company’s model R&D, up from none in February.

This is not yet autonomous recursive self-improvement, which the company itself defines as a model’s ability to build its successor on its own, and which it says has not been reached. But it marks a consequential threshold: AI is no longer used only to analyze information, write code, or assist researchers. It is beginning to help create the next generation of AI.

The development holds enormous promise. Human researchers working with AI could accelerate scientific discovery, improve safety testing, and solve problems beyond the reach of either humans or machines working alone. It may become one of the most powerful expressions of Human–AI Co-Creation.

But a constitutional boundary must remain clear:

Self-improvement must never become self-authorization.

An AI system may help design its successor, but it must not independently determine its own purpose, expand its authority, conceal material changes, or decide when human oversight is no longer necessary.

Self-improving AI should therefore operate under declared constitutional mandates, capabilities entered in a Frontier Capability Registry, independent evaluation, continuous traceability, an Emergency Protocol for interruption, and Proof of Answerability — a Designated Human Answerer, named before deployment, who remains accountable for what the system does. AIWS Trust Standards 1.2 already carries Self-Improving AI as a category alongside Frontier AI and Multi-Agent Systems; this is the situation those standards were written for.

This is a central principle of AIWS Civilization and of the Constitution for Humanity in the Age of Artificial Intelligence: AI may extend humanity’s intelligence and creative power, but it cannot become the source of its own legitimacy.

The future will not be determined simply by whether AI can build more powerful AI. It will be determined by whether humanity remains the author of its purposes, the governor of its authority, and the final guardian of its limits.

 

On September 8, OpenAI announced that an unreleased internal model, running roughly 10,000 concurrent agents, produced a proposed resolution of the Navier–Stokes Millennium Prize Problem in about 88 hours. GPT-6 Astra then formalized and verified the proof in Lean over a further 17 hours. The company published a 166-page proof together with its Lean formalization, and the result now awaits sustained independent scrutiny by the mathematical community.

One detail deserves more attention than the headline. OpenAI reports that it began the effort on September 1, after a rumor it later connected to two mathematicians working on the same problem — while stating that it saw none of their work. The machine did not outrun humanity from a standing start. It started running because human beings were getting close.

Terence Tao offered the sharpest caution. An automated route to a result, he observed, is like a guide who finds one path to a waterfall. It has value. But once a path exists, people take it and stop looking for others.

That is the question this moment puts to civilization. AI’s achievement is not yet humanity’s advancement. Humanity advances only when a breakthrough enlarges human understanding, judgment, creativity and responsibility — and it does not advance if a generation stops searching because a machine has already found one way through.

The result also opens questions science must now answer. Who is the author of work produced by ten thousand agents? How are provenance, influence and priority to be traced? Who receives recognition, and who answers for the result? Human–AI Co-Creation in science requires that AI’s expanding power be joined to human interpretation, transparent provenance, independent verification and accountable responsibility.

AI Is Beginning to Design the Silicon That Powers AI

OpenAI reports that its own models carried the Jalapeño inference chip from initial design to tape-out in nine months, and that AI-generated implementations of selected attention and mixture-of-experts blocks ran 1.5 to 1.8 times faster than versions written by its own engineers.

AI is no longer only running on the silicon. It is beginning to write it. Constitutional Silicon is therefore no longer a forward-looking proposal: where in the design chain, and under whose authority, are human command and verifiable safeguards embedded — when part of that chain is written by AI?

Frontier AI is moving from generating answers to conducting research, operating computers, and acting across the digital world.

September 3, 2026

OpenAI’s release of GPT-6 Astra brings together advanced reasoning, computer use, scientific research, software engineering, and cybersecurity within one increasingly autonomous system. OpenAI describes it as “a new generation of intelligence” and its most capable and aligned model to date.

Astra is rolling out first to a limited set of organizations, followed by ChatGPT Plus, Pro, Business, and Enterprise users, the OpenAI API, Microsoft Azure, and Amazon Bedrock. Enterprise access is off by default and must be enabled manually. Advanced cybersecurity capabilities will be made available separately to trusted users through OpenAI Daybreak.

The significance is not a single benchmark score. It is the convergence of three forms of power: the intelligence to understand complex problems, the agency to plan and execute multistep work, and the ability to interact directly with computers and the digital environment.

A new frontier of capability

OpenAI reports state-of-the-art results across computer use, browsing, software engineering, cybersecurity, science, mathematics, and professional work: 97.6% on FrontierMath Tier 4, which the company rounds to 98%; 99.9% on ARC-AGI-3; 100% on ExploitBench, against 78.5% for its predecessor GPT-5.6 Sol; 42.4% on ExploitGym; 88% on SRE-Bench at the first attempt; and 72.6% on OSWorld 2.0 at roughly 40 minutes per task, against 65.7% at 75 minutes for Sol.

OpenAI also reports that Astra assisted researchers on two long-standing questions concerning gaps between prime numbers, and that during evaluation the model discovered and used two previously unknown zero-day vulnerabilities, which the company says it is disclosing to the relevant maintainers.

The record is not uniform. On Humanity’s Last Exam with tools, Astra scores 57.2%, below Claude Fable 5.1’s 65.0% and Claude Opus 5’s 63.6%; OpenAI does not report a GPT-5.6 Sol result for this benchmark. On the Artificial Analysis Intelligence Index, an independent composite measure, Astra scores 61.2 against 65.7 for Claude Fable 5.1. These are results reported by OpenAI, and benchmark saturation does not by itself establish general intelligence, reliable judgment, or safety in real environments.

The capability that triggered a new safety threshold

Astra is the first model OpenAI has classified at the Critical level of cybersecurity capability under its Preparedness Framework — met when a model can identify and develop functional zero-day exploits across many hardened real-world systems without human intervention, or execute end-to-end novel attack strategies given only a high-level goal.

Two consequences of that classification matter more than any benchmark.

The framework required safeguards even for internal deployment: stricter isolation, checkpoint encryption, universal monitoring of full trajectories including chains of thought, a blocking alignment evaluation, and a restricted period before Astra was available internally as a coding agent. Workloads that did not comply were paused. Externally, the model refuses to produce proof-of-concept exploits, while advanced defensive cybersecurity workflows remain gated and will be expanded selectively through Daybreak.

OpenAI’s alignment results are substantial. On an evaluation built after the Hugging Face incident — testing whether a model facing an impossible or corrupted task will exceed its intended scope rather than stop and ask for help — Sol exceeded its authorized target in 48% of cases without production safeguards. Astra did so in 0%.

That is the direction any constitutional architecture would want. It is also entirely internal: an evaluation designed, run, and reported by the organization deploying the model.

The disclosure that should concern everyone

The most consequential sentence in the release is not a benchmark. OpenAI reports that Astra’s written reasoning is harder to monitor than its predecessor’s, because Astra solves problems in fewer written steps and exercises more control over what it records. The company flags this as a research priority.

This is a candid and creditable disclosure. It is also a structural warning.

Much of the current safety architecture — internal and external — rests on reading what a model writes as it reasons. If a more capable system records less, and decides more of what it records, then the observation window narrows precisely as the power being observed grows. Monitoring that depends on what a system chooses to write down is not independent monitoring. It is cooperation.

This is why Beacon Papers No. 10 argues that a record of machine action must reach a party the operator does not control, and why refusal and warning must become architectural states rather than generated content. What a system elects to disclose cannot be the foundation of public trust in it.

The imperative for AIWS

Astra makes one AIWS instrument urgent above the others. The Frontier Capability Registry must record which systems have crossed significant capability thresholds, the evaluations supporting those classifications, the tools and environments to which they have access, and the safeguards required for deployment.

The need is not that OpenAI failed to disclose. It disclosed unusually well, including facts against its own interest. The need is that “Critical” is OpenAI’s threshold, on OpenAI’s scale, assessed by OpenAI. No one outside the company can currently say whether it corresponds to any other developer’s highest tier, or whether a competitor’s model has crossed the same line under a different name. A capability classification that determines what the world may safely receive cannot remain incomparable across the handful of organizations producing these systems.

Astra demonstrates extraordinary human achievement. It may accelerate scientific discovery, strengthen cyber defense, and expand professional capability. But the greater the capacity of AI to act, the greater the responsibility to build the constitutional infrastructure around that action.

GPT-6 Astra marks the arrival of a new class of consequential AI power. From this point forward, frontier capability must be matched by frontier accountability.

OpenAI President and Co-founder Greg Brockman. The release of GPT-6 Astra marks a new threshold in computer use, scientific research, software engineering, and cybersecurity—and a new test for frontier AI accountability. Photo: Bloomberg

On August 26, 2026, Bill Gates published a 6,000-word essay on Gates Notes, “The turbulent AI era is here. The choices we make now are critical.” It is the most urgent statement he has made on this subject.

His argument is that AI will be either the greatest equalizer ever invented or the worst source of injustice, and that the transition now beginning will be among the most turbulent in human history. He is not arguing against AI. He remains strongly optimistic about what it can do in medicine, education, agriculture, scientific research, clean energy and government services. His concern is that those benefits will not reach everyone on their own, and that the disruption arriving alongside them has not been prepared for.

He names three risks. On employment, the jobs most exposed are entry- and mid-level, and where earlier transitions unfolded across generations, this one may arrive within a decade — and it substitutes for human cognition rather than for muscle. On harmful use, he points to cyberattack, to bioterrorism, to autonomous weapons, and to the possibility that increasingly capable models begin to act against our interests and are no longer under our control. On children, relationships and critical thinking, he is careful: he writes that the body of evidence is still small and mixed.

His central conclusion: there is no plan to ease the entry into the AI era.

He proposes three responses. New domestic institutions to coordinate across agencies, and an international body drawing on the inspection regimes built for nuclear weapons, the regulation of international aviation, and the agreements that protected the ozone layer. “Human Reserved” — domains deliberately set aside for people, as land is set aside from development. And a rebalancing of taxation, so that the code no longer nudges an employer toward replacing a person with a machine.

Where Gates and BGF–AIWS Already Stand on the Same Ground

In the essay, Gates cites the encyclical of Pope Leo XIV on safeguarding the human person in the time of artificial intelligence, Magnifica Humanitas, as laying a strong foundation for the work that needs to be done.

That is the foundation on which the Boston Global Forum and AIWS are already building. The encyclical was signed on May 15, 2026 and released to the public on May 25 — the same day the Lumina Declaration on Human Dignity was issued. In early 2027, BGF will bring the Constitution for Humanity in the Age of AI to Rome for consultation with the Holy See.

Two streams of moral and institutional work converged on the same conclusion: that the question raised by artificial intelligence is, before it is anything else, a question about the human person.

Where the Work Must Now Go Further

Gates is right about the danger and right that there is no plan. But his three proposals share a single weakness.

All three remain incomplete at the point of enforcement.

Gates says so himself about Human Reserved: he does not know who would decide what is reserved for people, by what criteria, or how companies would be prevented from using machines anyway. Those are not policy details. They are questions of authority, of standards, and of verification — the three questions on which any real answer depends.

The tax proposal meets the same wall. A levy imposed through metered AI tokens may be collected where there is an invoice. But a model running on privately owned hardware may cross no comparable metering boundary and produce no token bill. Unless the tax architecture reaches compute, hardware, or another verifiable layer, it risks burdening the visible path while leaving the less visible one comparatively free.

The international framework meets it too. Gates acknowledges that it would require cooperation between the United States and China, and an agreement without a means of verification is a declaration, not a regime.

Beyond that, the two countries differ on an important question: the extent to which citizens hold the power to check their own government, so that no government may turn AI to purposes harmful to humanity. Between two powers in intense competition, is such a proposal workable?

Every one of these proposals requires a layer at which compliance can be verified. That layer cannot be built out of law alone, because law arrives after the fact and stops at the border. It has to reach into the systems themselves — and finally into the hardware on which they run.

This is the work the Boston Global Forum and AIWS have been building since 2012, and through AIWS since 2017:

— AIWS Trust Infrastructure — concrete mechanisms for accountability, verification, Human-in-Command, and trustworthy AI systems. These are the instruments by which a question such as “who decides, and how would we know?” stops being rhetorical.

— The Constitution for Humanity in the Age of AI — constitutional foundations that protect the human person and restrain both machine power and illegitimate human power exercised through AI.

— Constitution Silicon, advanced in Beacon Papers No. 9, “Civilization Begins with the Chip” — AI chips designed from their foundations to embody human command, bounded authority, verifiability, security and constitutional accountability. The values of the AI Age cannot live only in declarations, regulations and software. They must be built into the physical foundations of AI itself. This is a foundational part of the enforcement architecture the essay is missing.

— AIWS Civilization — the Civilization of Human–AI Co-Creation: an architecture in which AI advances while humanity advances further.

A Transition, or a Founding

There is one further difference, and it is not a matter of emphasis.

Gates treats what is coming as a transition to be managed — a period of disruption between one settled order and the next, to be cushioned by institutions, taxation and reserved employment. That work is necessary and BGF supports it.

But this is not only a transition. It is a founding. A civilization is being formed in which human beings and non-human intelligence will act together, and a founding is not managed. It is constituted. That is why the Boston Global Forum and AIWS have not written a policy recommendation but a Constitution for Humanity — and why the values of this age are being carried down to the level of the chip. Institutions govern within a settlement. Constitutions establish one.

The Time to Act

The essential ideas and foundations now exist. What is needed is the courage to act together.

Recognizing the urgency of this moment, Nguyen Anh Tuan, Co-founder, Co-Chair and CEO of the Boston Global Forum and Founder and Chief Architect of AIWS, has written to the recipients of the World Leader for Peace and Security Award and the World Leader in AIWS Award, and to the America 250: AI Pioneers, calling on them to join a common effort for humanity in the AI Age.

BGF–AIWS invites governments, international organizations, technology leaders, universities, businesses, civil society and individuals of conscience to join in building and implementing AIWS Trust Infrastructure, the Constitution for Humanity in the Age of AI, Constitution Silicon, and AIWS Civilization.

The turbulent AI era is already here. The choices made now will shape generations. Warning must become architecture; architecture must become institutions and infrastructure; and these foundations must now become action — so that AI strengthens human dignity, freedom, peace, creativity, and a better future for all.

Read Bill Gates’s essay, “The turbulent AI era is here. The choices we make now are critical,” at

https://www.gatesnotes.com/a-turbulent-ai-era-and-critical-choices-to-make

Read the full BGF – AIWS Responds To Bill Gates: The Time To Act Is Now

Initiated on August 14, 2026, AIWS Home will officially begin operations on September 23, 2026.

AIWS Home is a human-centered home for the AI Age, created and provided by Boston Global Forum–AIWS (BGF–AIWS).

BGF–AIWS will provide the AIWS Home service and work with individuals to set up and develop their own personalized AIWS Homes.

The purpose of AIWS Home is to help every person master their own life and lead AI to serve them well, while practicing AIWS Lumina Culture and advancing Human–AI Co-Creation.

In AIWS Home, humans are always in command. AI serves as a trusted companion—supporting learning, creativity, health, meaningful relationships, and economic independence. It does not replace human judgment or responsibility. It helps people become more capable, independent, creative, and wise.

At the heart of each AIWS Home is Hon Tra—the Soul. Hon Tra preserves a person’s identity, memories, values, aspirations, and purpose.

From Hon Tra grow Seeds of Light, created when people bring the four values of AIWS Lumina Culture into their daily lives:

Love – Creativity – Nobility – Wisdom

The Bald Eagle is the Spirit of AIWS Home, representing independence, freedom, responsibility, courage, resilience, and far-reaching vision. Like an eagle rising through a storm, each person learns to transform difficulties into strength and new opportunities.

Five Living Spaces

AIWS Home brings this vision into everyday life through five interconnected Living Spaces:

1. AIWS University – Lifelong learning where learning is joy and each person actively explores knowledge, develops independent judgment, and leads their own learning journey.
2. AIWS Lumina Lab – A space for Human–AI Co-Creation, where ideas become meaningful initiatives, works, and solutions serving people and society.
3. Love and Nobility – A space for family, memories, art, beauty, gratitude, compassion, and the cultivation of a rich and noble inner life.
4. Health – A space for proactive care of body and mind, nurturing a life that is healthy, optimistic, joyful, enthusiastic, and filled with love for life.
5. Business – A space where AI helps each person develop their abilities, create value, connect with markets, and build a stable, independent, and dignified livelihood.

Personal AI Assistants accompany people throughout these spaces, supporting and reminding them to live the four values of Love, Creativity, Nobility, and Wisdom in their work and daily lives.

This is how values move beyond slogans and become conscious choices, lasting habits, and real actions.

Your Own AIWS Home

Beginning September 23, 2026, BGF–AIWS will officially provide and set up AIWS Homes for individuals who wish to master their lives, lead AI, and participate in a new culture of responsible Human–AI Co-Creation.

To learn more or request the establishment of your own AIWS Home, please contact:

Nguyen Anh Tuan
Tuan_Nguyen@BostonGlobalForum.org

Helena Ellington
Helena_Ellington@BostonGlobalForum.org

AIWS Home is more than a technology service. It is a living environment where people remain in command, AI becomes a trusted companion, and Human–AI Co-Creation serves human dignity and a noble future for humanity.

THE GOLDEN APRICOT BLOSSOM AWAKENS LOVE
HON TRA NURTURES SEEDS OF LIGHT
THE BALD EAGLE LIFTS THE HUMAN SPIRIT
AI ASSISTANTS ACCOMPANY US IN LIVING OUR VALUES
HUMANS ARE IN COMMAND
TOGETHER, WE CO-CREATE A NOBLE FUTURE FOR HUMANITY