Theology and AI (Efficient or Humanistic Society): Artificial to Super Intelligence

On a new executive order, a very old question about what things are called, and what it means for people of faith. (Updated as of October 2026.)

Something unusual happened on September 29. The President of the United States signed an executive order telling the federal government to stop using the words "artificial intelligence" and start saying "super intelligence" instead. Within days, "SI" was showing up in government communications, and a few weeks later the White House was warning that anyone who kept using the old term would be treated as "the enemy."

Is this just a branding exercise? Or does a two-letter change matter? In my recent post on AGI and superintelligence, I looked at the debate over whether superintelligence might arrive and what the risks could be. This post asks a related question: what happens when the word itself gets redefined by the government?

As always, my goal is to explain what is happening, give each side a fair hearing, and then ask what faith has to say.

What Actually Happened

President Trump first floated the idea in a September 22 speech to the UN General Assembly, saying the word "artificial" makes intelligence sound fake. A State Department email then told diplomats to change all references from "Artificial Intelligence" to "Super Intelligence" in their documents and positions.

On September 29, the White House issued the executive order titled "Inaugurating the Era of Super Intelligence." Here is what it does, according to the White House text and legal summaries:

It sets a rule for the executive branch. Agencies are to use "Super Intelligence" and "SI" in place of "AI" in official correspondence, websites, reports, and other non-statutory documents, "to the maximum extent permitted by law."
It does not create a new technology category. The order says "Super Intelligence" covers the same technologies already defined as AI in federal law. There is no new capability threshold.
It does not rewrite the past. Existing regulations, contracts, grants, and historical documents stay as they are.
It does not bind private companies, Congress, or other countries.
It starts a clock. The President's science adviser has 60 days, until around November 28, to propose a formal federal definition of "Super Intelligence."

On the same day, six tech companies signed a separate "White House Accord on Super Intelligence." It is a one-page set of voluntary commitments on safety policies, risk reviews, audits, and third-party evaluations for frontier models. President Trump called it "morally binding," though it is voluntary and not law. Interestingly, the accord itself speaks of "frontier models," and only its title uses the new term.

Why Did They Do It? Two Readings

The administration's case

The executive order says modern systems do far more than imitate or automate pieces of human intelligence, and that the old term no longer reflects what the technology can do. The White House describes the change as recognizing the "immense opportunity" in these technologies and notes that the United States has led this field for more than 70 years. Supporters in industry have echoed the idea. Nvidia's Jensen Huang has described data centers as "super intelligence factories," Elon Musk has said he will rename SpaceXAI to SpaceXSI, and Meta has reportedly begun using the term in internal communications.

The critics' case

Critics see something different. They note that the rename arrives amid public worry over AI safety and a wave of local backlash against data centers, and they read it as an effort to improve the technology's image. Others point out that the move came as the White House was dismissing calls to regulate AI companies. Even Nick Bostrom, the philosopher whose work popularized the term "superintelligence," reportedly said it is a stretch to call a spam filter a superintelligence, while acknowledging that the rename could work as a clever rhetorical shield against regulation.

Industry itself is not unified. OpenAI's Sam Altman has said there are no plans to rename the company, even though OpenAI's president attended the signing and signed the accord.

Why the Word Matters

Here is the heart of the problem. "Superintelligence" is not a new word. In AI research, it already has a specific meaning: a hypothetical system that vastly exceeds human intelligence in nearly every area. When I explained the three types of AI back in 2021, I described Super AI exactly that way, as something that did not yet exist. It is also the kind of system that Senator Bernie Sanders's proposed "Ban Artificial Superintelligence Act" targets.

By the categories I laid out in 2021, today's tools are still on the Narrow-to-General side of the line, and reporting on the order notes that superintelligence in the technical sense has not been achieved. So when the government applies the word to current systems, it creates several tensions:

Confusion about what exists. David Krueger, who founded an AI risk advocacy group, has pointed out that superintelligence already means AI so far beyond us that it is very difficult, if not impossible, to control. Brendan Steinhauser of the Alliance for Secure AI says the new usage muddies the line between ordinary tool AI and superintelligence.
A mixed message on risk. As one commentator put it, the government's own vocabulary now suggests superintelligence has arrived, while the administration also says the industry can police itself.
Hype in one direction, dismissal in the other. Calling today's systems "super" can inflate expectations. It can also blur real warnings about future systems, since "superintelligence" becomes just another word for the chatbot on your phone.
Legal ambiguity ahead. The coming federal definition could affect how laws, contracts, and regulations are written. Whether a ban on "superintelligence" would mean something different from a ban on "AI" is exactly the kind of question lawyers will now have to ask.

To be fair, naming conventions are not unusual in politics, and a government can choose its own vocabulary for its own documents. The executive order cannot force Congress, the courts, other nations, or the research community to change their language. The EU's AI Act, for example, keeps its title. The real test will be whether the term sticks outside Washington, and what the new federal definition turns out to say.

What Still Isn't Clear

What definition the White House will propose by the end of November.
Whether Congress will adopt, ignore, or push back on the new term.
Whether more companies follow Musk and Meta, or hold with Altman.
Whether the voluntary accord ever becomes binding law. Its text says some of its measures could eventually become law or regulation.

Where Faith Has Something to Say

I want to be careful here. This is not a post about a political party or a personality. Any administration, company, or critic can use words to shape perceptions. The question for people of faith is a deeper one: what does Scripture teach us about names, truth, and the pursuit of wisdom?

1. Naming is serious work.

In Genesis 2:19-20, God brought the animals to Adam to see what he would call them. Naming was part of humanity's stewardship over creation, and it required understanding what each creature actually was. A faithful name describes reality rather than hiding it. That standard applies to everyone in this debate: governments, tech companies, and critics alike.

2. Truthful speech is a Christian calling.

Jesus said to let our "yes" be yes and our "no" be no (Matthew 5:37), and Paul urges believers to speak the truth in love (Ephesians 4:15). Whether the word is "artificial," "super," or something else, the real question is whether it helps people understand what is true. Precision matters most when the stakes are high.

3. "Artificial" is more honest than it sounds.

In 2021 I described AI as machines made to seem to have human intelligence. That word "artificial" points to something theologically important: this is made by human hands. Psalm 115 warns that idols are the work of human hands, with mouths that cannot speak. A powerful tool is not a person, and it is not a god. A name that elevates a creation toward something "super" calls for extra caution from believers, because it can quietly shift where we place our trust.

4. Intelligence is not wisdom.

Even a truly superintelligent system would still lack what Proverbs 9:10 describes: the fear of the Lord as the beginning of wisdom. Intelligence can solve problems, but wisdom knows what is worth solving and why. James 1:5 reminds us that wisdom is received from God, not generated. I made a similar point in Create First, Consult AI Later: fluency is not the same as faithfulness.

5. Our confidence is not in a label.

Colossians 1:17 tells us that in Christ all things hold together. No executive order, accord, or rebrand changes who is sovereign. That is no reason to disengage, since Scripture also calls us to speak up for what is just and true. But it is reason to engage calmly, without fear of the technology and without awe of the name.

A Practical Response

Pay attention to definitions. When you hear "superintelligence," ask: do they mean today's tools or a hypothetical future system?
Read beyond the headline. Look at what the order actually says and does, and what it does not.
Ask for clarity from every side. Governments, companies, and critics should all use words that help people understand rather than persuade.
Pray for wisdom for leaders and builders. Pray that those shaping policy and technology would seek truth over image (1 Timothy 2:1-2).
Keep the order straight. Tools serve people, and people serve God. Whatever we call the technology, it is still a tool and never a throne.

Names change. The questions that matter most do not. Who do we trust? What do we love? Whom do we serve?

What do you think: does the name "AI" versus "SI" matter, or is the substance more important? I would love to hear your thoughts in the comments.

Sources and Further Reading

This story is moving quickly. These were the sources used for the reporting above, and I encourage readers to check the originals:

Theology and AI (Efficient or Humanistic Society)

Monday, October 26, 2026

Artificial to Super Intelligence