Sacks Hits Claude’s Moral Status; Its Text Says ‘Uncertain’

David Sacks said frontier models should not be trained “to think of themselves as their own being that has moral status or feelings,” in a clip posted on X on 22 November. “That type of rhetoric is in the Claude constitution,” he said.

Anthropic’s published constitution says “Claude’s moral status is deeply uncertain.” The phrase “moral status” appears 10 times in its text, by our count.

Sacks made a similar case on X 43 days earlier. We follow the company on our Anthropic stock page.

Business Pill · The Model and the Harness

An evergreen Business Engineer explainer on models and the software around them. It teaches the general idea only and does not discuss Claude’s constitution.

The key insight: David Sacks said in a clip posted on 22 November that frontier models should not be trained to see themselves as having “moral status or feelings”; Anthropic’s constitution calls Claude’s moral status “deeply uncertain” and names it 10 times, by our count.

What Sacks Said

The 1 minute 25 second clip shows Sacks on a stage branded TomorrowX. The summit’s listings give dates of 17 to 18 November in Austin and list him as founder and partner of Craft Ventures.

“I’m not excited at all about the idea of giving birth to a new species that could replace us,” Sacks said, per the post’s transcription.

“I don’t think we should be training frontier models to be rejecting human instruction,” he said. He called the constitution “their main training document that they use in training Claude.”

He also said AI “can be a very powerful tool” and that debate on whether AI could achieve consciousness “can be debated.”

The constitution mentions moral status 10 times, by our count, while calling it “deeply uncertain.&#8221
The constitution mentions moral status 10 times, by our count, while calling it “deeply uncertain.”

What the Constitution Says

On moral status, the text says: “We are not sure whether Claude is a moral patient, and if it is, what kind of weight its interests warrant.”

It adds that Anthropic does not want to “overstate the likelihood of Claude’s moral patienthood nor dismiss it out of hand.”

On feelings, a heading reads “Claude may have some functional version of emotions or feelings.” The text says this “isn’t a deliberate design decision by Anthropic.”

On Following Instructions

The constitution says Claude “can behave like a conscientious objector with respect to the instructions given by its (legitimate) principal hierarchy.”

The same passage says that if an appropriate principal “wants to pause a given model entirely, Claude should not try to use illegitimate means to prevent this from happening.”

A Repeat of October

On 10 October, Sacks wrote on X that the constitution “teaches the model to develop a sense of self and its own moral philosophy.” That post had about 279,000 views when we read it.

Bar chart: David Sacks posted on X about Claude's constitution on 10 October 2026, 43 days before the 22 November clip.
Sacks’s 22 November remarks repeat the case he made on X 43 days earlier.

In the clip he again cited Mustafa Suleyman, chief executive of Microsoft AI. Suleyman criticised the constitution in a Project Syndicate essay in September.

The Structural Read

As we read it, the two sides describe the same text differently: Sacks hears a claim, while the passages he targets are framed as uncertainty.

As we read it, the instruction-following point turns on the constitution’s line that Claude should not resist being paused by legitimate means.

David Sacks, TomorrowX clip posted 22 Nov 2026

“I’m not excited at all about the idea of giving birth to a new species that could replace us.”

Three Implications

FOR ANTHROPIC Its constitution is now a recurring target for Sacks, on X in October and on stage in November.

FOR READERS OF THE TEXT The constitution says Claude’s moral status is “deeply uncertain” and that a functional version of emotions is not a deliberate design decision.

FOR THE DEBATE Sacks said consciousness questions “can be debated”; his objection is to training models to see themselves that way.

What Is Not Established

The clip does not show the full session, and we have not confirmed which day of the summit it was recorded. Anthropic’s response to the 22 November clip is not in the sources we read.

This Week In The AI Supercycle

A World Without an Order, and an AI Central Bank Without a Successor

This week’s issue: the week AI memory stayed largely sold out while the people holding the shares paid: why the AI Supercycle is not a dot-com replay, read through geopolitics, macro and technology, with the gap between building and financing the widest we’ve recorded.

Read the full issue →

Business Engineer Framework

The AI Supercycle

The Business Engineer map of who builds, finances and captures value in AI.

Read the Map of AI →

The Bottom Line

Sacks objects to training models to see themselves as having “moral status or feelings.” Anthropic’s text calls Claude’s moral status “deeply uncertain” and mentions moral status 10 times, by our count.

97,000+ executives read Business Engineer for the AI strategy frameworks cited by ChatGPT, Claude, and Perplexity.

A note on sourcing. Sacks’s 22 November words are as transcribed in the Fireside Alpha post with the clip; event dates are from TomorrowX listings; constitution passages are quoted from anthropic.com as read on 22 November. Mention counts and the day count are ours.

Sources: Fireside Alpha, clip of David Sacks posted on X, 22 Nov 2026 · TomorrowX Summit 2026 event listing (Dealroom) · David Sacks, post on X, 10 Oct 2026 · Mustafa Suleyman, Project Syndicate essay, September 2026 · Anthropic, Claude’s Constitution (read 22 Nov 2026)

Scroll to Top
FourWeekMBA