Chinese Quantum Test Confirms Feynman Math, Not Physical Reality

Sabine Hossenfelder

Tracking photons across intermediate measurement points verifies standard quantum probability calculations but cannot prove that unobserved particles physically travel every path simultaneously.

AI-assisted summary of FINALLY: A Proof that Particles Take All Paths At Once? — original source. Check the original for full context.

Source published: 2026-09-24

Key ideas

Physicists in China published experimental results claiming to prove that Richard Feynman's path integral represents physical reality rather than a mere mathematical tool. (0:00)
The experimental design sent single photons through a spatial grid of 17 longitudinal by 5 transverse points to track amplitude evolution between intermediate locations. (3:15)
Sabine Hossenfelder disputes the authors' interpretations, rating the claim 9 out of 10 on the 'bullshit meter' because measuring intermediate points cannot reveal what particles do unobserved. (3:15)
The experiment fails to bypass foundational quantum limitations formulated by Einstein and Bell, leaving quantum interpretation questions unresolved. (4:33)

Detailed notes

## Background Published on YouTube by Sabine Hossenfelder on September 24, 2026, this video examines sensational claims in quantum physics. According to the description, Sabine evaluates a study by a Chinese research team asserting they experimentally proved Richard Feynman's eighty-year-old postulate that particles physically take all paths simultaneously.

## TL;DR Sabine analyzes an experiment claiming to prove that quantum particles physically travel every possible path at once, rather than treating Feynman's path integral merely as a mathematical calculation tool. She explains that Feynman's framework sums amplitudes over unobserved trajectories, but the Chinese researchers introduced a physical grid of intermediate measurement points. By measuring the photons along these points, the team simply confirmed Feynman's uncontested mathematical formulas rather than discovering what particles do when unobserved. Consequently, Sabine awards the paper a nine out of ten on her 'bullshit meter' and notes that unresolved foundational questions raised by Albert Einstein and John Bell remain entirely untouched.

## Atmosphere & style The video has a sharp, wry, and no-nonsense tone, driven by Sabine's signature deadpan delivery and impatience with scientific sensationalism. She blends dry, self-deprecating humor with rigorous analytical precision, taking evident pleasure in dismantling overhyped press headlines. Watching it feels like a grounding, clarifying masterclass in separating legitimate optical experimental results from unwarranted philosophical hype.

## Bulletin - [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) Physicists in China published experimental results claiming to prove that Richard Feynman's path integral represents physical reality rather than a mere mathematical tool. - [03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) The experimental design sent single photons through a spatial grid of 17 longitudinal by 5 transverse points to track amplitude evolution between intermediate locations. - [03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) Sabine Hossenfelder disputes the authors' interpretations, rating the claim 9 out of 10 on the 'bullshit meter' because measuring intermediate points cannot reveal what particles do unobserved. - [04:33](https://www.youtube.com/watch?v=A2ImadKWaHA&t=273s) The experiment fails to bypass foundational quantum limitations formulated by Einstein and Bell, leaving quantum interpretation questions unresolved.

## Key takeaways - [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) Classical mechanics predicts single deterministic trajectories, whereas quantum mechanics calculates probabilities across all conceivable paths. - [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) Richard Feynman introduced the path integral formulation 80 years ago, summing amplitudes over all possible paths to yield probabilities matching Schrödinger's wave equation. - [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) Feynman's formulation produces a vivid mental picture of particles taking all paths simultaneously, often used to justify parallel universe interpretations. - [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) Google Quantum AI previously cited path-like quantum behaviors to suggest quantum computers operate across parallel multiverses. - [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) Path integrals inherently compute paths that are never directly observed, because placing intermediate detectors collapses the system to a single observed path. - [03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) The Chinese research team sent single photons through a spatial grid of 17 longitudinal points and 5 transverse points. - [03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) The experiment verified that Feynman's mathematical formulas correctly calculate amplitude evolution between specific measurement points. - [03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) The experimental setup did not demonstrate unobserved behavior, but instead simply detected photons traversing observed points. - [03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) Hossenfelder rates the paper's interpretive claim a 9 out of 10 on the bullshit meter despite the quality of the optical execution. - [04:33](https://www.youtube.com/watch?v=A2ImadKWaHA&t=273s) The debate reflects the long-standing question explored by Einstein and Bell regarding whether unobserved systems have definite hidden states. - [04:33](https://www.youtube.com/watch?v=A2ImadKWaHA&t=273s) Bell's theorem rules out local realism only if superdeterminism is assumed to be false, an assumption that remains untested. - [04:33](https://www.youtube.com/watch?v=A2ImadKWaHA&t=273s) Intermediate path measurements cannot resolve fundamental quantum interpretation problems, leaving the foundational interpretation of quantum physics an open question.

## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) Feynman's Path Integral and the Multiverse Debate

In classical mechanics, macroscopic objects follow deterministic, predictable trajectories. If you throw a ball, its trajectory is dictated entirely by its initial position and velocity—or as Sabine jokes, if she throws it, she will likely tear an unfamiliar body part while watching the ball vanish into the bushes. In quantum mechanics, however, this deterministic pathing vanishes. Eighty years ago, Richard Feynman proposed the path integral formulation: rather than calculating a single trajectory connecting a starting point to an endpoint, one considers every conceivable path connecting the two points. An amplitude is computed for every possible path, and these amplitudes are summed together to determine the overall probability of finding the particle at a specific destination.

```mermaid flowchart TD accTitle: Comparison of Classical Path versus Feynman Path Integral accDescr: Classical mechanics follows a single deterministic trajectory while quantum mechanics sums amplitudes across all paths. subgraph Classical["Classical Physics"] C1["Initial State"] -->|"Single deterministic path"| C2["Final State"] end subgraph Quantum["Feynman Path Integral"] Q1["Source"] -->|"Path A amplitude"| SUM{"Sum Amplitudes"} Q1 -->|"Path B amplitude"| SUM Q1 -->|"Path C amplitude"| SUM SUM -->|"Calculate probability"| Q2["Detector"] end ```

Feynman proved that this path integral formulation yields the exact same mathematical predictions as the Schrödinger equation using the wave function. The conceptual appeal of the method is that it offers a vivid mental picture of quantum weirdness: particles seemingly taking all possible paths at the same time. This conceptual model has spurred decades of foundational debate. Many physicists treat this description as literal reality, sometimes citing it as evidence for parallel universes. Sabine points to a Google quantum computing press release claiming their experiments gave "credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch."

- The fundamental measurement catch: by construction, path integrals sum over unobserved paths. An observer only detects the particle at its destination; placing an intermediate detector collapses the system into observing only a single path. - A research team in China recently published a paper asserting that these paths are "more than mere mathematical artifacts" and actually "reflect a physical aspect of quantum reality."

---

[03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) What the Chinese Experiment Actually Measured

The validity of Feynman’s mathematical framework is not under dispute—it has been verified by thousands of experiments over several decades. Rather, the crux of the issue is whether the Chinese team's experiment demonstrated what particles do when they are *not* being watched. In their setup, the researchers sent individual photons, one at a time, through a spatial grid of measurement locations consisting of 17 points along the propagation axis and 5 points transverse to it. By picking pairs of grid points, they tracked the amplitude evolution between them and confirmed that the mathematical behavior matches Feynman’s path integral predictions.

Headlines quickly followed, with *New Scientist* declaring: "Richard Feynman’s 80-year-old quantum postulate has now been validated." While Sabine notes that the *New Scientist* headline is actually more measured than the authors' own conclusions, the claim of proving physical reality falls flat:

- The experiment validated that Feynman's mathematical formulas add up correctly between points of interaction, a result no physicist doubted. - It did not determine what photons do in the absence of observation. Instead, by inserting intermediate grid measurements, the experiment simply detected photons traversing specific measured paths. - Sabine awards the paper a "9 out of 10 on the bullshit meter," characterizing it as a well-executed optical experiment hitched to a deeply flawed physical interpretation.

---

[04:33](https://www.youtube.com/watch?v=A2ImadKWaHA&t=273s) Foundations of Physics, Bell's Inequality, and Upgrading Coding Skills

This conflation highlights why progress in the foundations of physics frequently stalls. The team attempted to resolve the identical dilemma that Albert Einstein and John Bell grappled with: does a quantum system truly occupy all potential configurations when unobserved, or does it hold a definite state that remains concealed? Bell's inequality was designed to interrogate precisely this question. Bell's theorem only forces a non-local or non-realist interpretation if one assumes superdeterminism is impossible—an assumption that has never been definitively ruled out. Setting up another path measurement cannot bypass this fundamental limitation. As Sabine concludes: "Feynman is still right, we still don’t know how to interpret quantum physics, and John Bell is still dead."

Shifting gears to personal projects, Sabine mentions that the last time she wrote software it was in Fortran, prompting her to modernize her backend skills using the video's sponsor, Boot.dev.

- Boot.dev focuses on backend languages that power real-world infrastructure, such as Python, SQL, and Go. - Rather than passive video consumption, the platform uses an interactive, browser-based coding setup that mimics an RPG progression system alongside an endless practice area called the "training grounds." - Sabine highlights using it to better understand web security vulnerabilities and data handling while developing a personal platform for her content, offering viewers 25% off their first year using the code `SABINE`.

## Notable quotes > [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch. — Sabine Hossenfelder

> [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) more than mere mathematical artifacts — Sabine Hossenfelder

> [00:00](https://www.youtube.com/watch?v=A2ImadKWaHA&t=0s) reflect a physical aspect of quantum reality. — Sabine Hossenfelder

> [03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) Richard Feynman’s 80-year-old quantum postulate has now been validated. — New Scientist

> [03:15](https://www.youtube.com/watch?v=A2ImadKWaHA&t=195s) 9 out of 10 on the bullshit meter — Sabine Hossenfelder

> [04:33](https://www.youtube.com/watch?v=A2ImadKWaHA&t=273s) Feynman is still right, we still don’t know how to interpret quantum physics, and John Bell is still dead. — Sabine Hossenfelder

## Mentioned Richard Feynman, Erwin Schrödinger, Albert Einstein, John Bell, David Deutsch, Google, New Scientist, Boot.dev, Python, SQL, Go, Fortran

## Video description Click this link https://boot.dev/?promo=SABINE and use my code SABINE to get 25% off your first payment for boot.dev."

Eighty years ago Richard Feynman said that a quantum particle takes every possible path at the same time. A team of physicists in China now says they have experimentally shown this is not just an interpretation but physical reality, and the headlines duly followed. I had a look at what they actually measured.

Paper: DOI: 10.1126/sciadv.aeh1011

#science #sciencenews #physics #quantumphysics

👕T-shirts, mugs, posters and more: ➜ https://sabines-store.dashery.com/ 💌 Support me on Donorbox ➜ https://donorbox.org/swtg 👉 Transcript with links to references on Patreon ➜ https://www.patreon.com/Sabine 📝 Transcripts and written news on Substack ➜ https://sciencewtg.substack.com/ 📩 Free weekly science newsletter ➜ https://sabinehossenfelder.com/newsletter/ 👂 Audio only podcast ➜ https://open.spotify.com/show/0MkNfXlKnMPEUMEeKQYmYC 🔗 Join this channel to get access to perks ➜ https://www.youtube.com/channel/UC1yNl2E66ZzKApQdRuTQ4tw/join 📚 Buy my book ➜ https://amzn.to/3HSAWJW

[Source](https://www.youtube.com/watch?v=A2ImadKWaHA)

中文简报

中国量子实验证实费曼数学计算，而非物理现实通过中间测量点追踪光子验证了标准量子概率计算，但无法证明未被观测的粒子在物理上同时走过所有路径。 Sabine分析了一项声称证明量子粒子在物理上同时走过所有可能路径的实验，该实验并非仅将费曼路径积分视为一种数学计算工具。她解释称，费曼的理论框架是对未被观测的轨迹求振幅和，但中国研究人员引入了一个由中间测量点组成的物理网格。通过沿这些点测量光子，该团队仅仅证实了费曼毫无争议的数学公式，而并未发现粒子在未被观测时究竟做了什么。因此，Sabine在她的“胡扯指数”上给这篇论文打了9分（满分10分），并指出阿尔伯特·爱因斯坦和约翰·贝尔提出的悬而未决的基本问题依然完全没有得到解决。中国物理学家发表实验结果，声称证明理查德·费曼的路径积分代表了物理现实，而不仅仅是一种数学工具。 (0:00) 该实验设计让单光子穿过一个纵向17个点、横向5个点的空间网格，以追踪中间位置之间的振幅演变。 (3:15) Sabine Hossenfelder对作者的解读提出异议，并在“胡扯指数”上给该主张评了9分（满分10分），因为测量中间点无法揭示粒子在未被观测时的行为。 (3:15) 该实验未能绕过爱因斯坦和贝尔提出的基本量子局限，量子诠释问题仍未得到解决。 (4:33)

Reading summary

Sabine Hossenfelder argues that recent claims alleging experimental proof that particles physically take all paths at once misinterpret standard quantum measurements, conflating verified mathematical calculations with unobserved physical reality. Feynman's Path Integral and Quantum Trajectories (0:00) In classical mechanics, macroscopic objects follow deterministic, predictable trajectories governed by initial position and velocity. In contrast, quantum mechanics replaces single trajectories with Richard Feynman’s 80-year-old path integral formulation, in which amplitudes for every conceivable path connecting two points are computed and summed to determine the probability of finding a particle at a destination. Mathematically, this formulation produces the exact same predictions as the Schrödinger wave equation. While Feynman’s approach offers an intuitive mental model of particles traversing all paths at once, it has fueled persistent foundational disputes. Proponents have occasionally cited the formulation as physical evidence for parallel universes, including a Google quantum computing press release claiming their experiments lent credence to David Deutsch's multiverse hypothesis. However, Hossenfelder highlights a fundamental constraint: path integrals mathematically sum across unobserved trajectories, and any physical measurement introduced along the way collapses the system into a single observed path. Experimental Measurements Versus Unobserved Reality (3:15) A research team in China recently published an experiment asserting that quantum paths are more than mathematical artifacts and reflect a physical aspect of quantum reality, leading New Scientist to report that Feynman’s 80-year-old postulate had been validated. The setup sent individual photons through a spatial grid of measurement locations consisting of 17 points along the propagation axis and 5 points transverse to it, tracking amplitude evolution across selected pairs of points to confirm agreement with Feynman's formulas. Hossenfelder contends that while the optical execution was sound, the authors' physical interpretation is deeply flawed. The experiment validated that Feynman’s equations accurately predict transitions between interaction points—a mathematical fact not in dispute—but it could not demonstrate what photons do when they are not being observed. Because intermediate grid points directly measured the photons along specific paths, the experiment observed interaction points rather than unobserved quantum trajectories. Unresolved Interpretations and Theoretical Constraints (4:33) The conflation of mathematical tools with physical reality mirrors long-standing debates explored by Albert Einstein and John Bell regarding whether unobserved quantum systems occupy all configurations or possess definite, hidden states. Bell’s inequality was designed to test these distinctions, but Bell's theorem only compels a non-local or non-realist interpretation if superdeterminism is assumed to be impossible—an assumption that remains unproven. Hossenfelder stresses that adding intermediate path measurements cannot circumvent these foundational boundaries. Consequently, testing Feynman’s mathematics within a detector grid leaves quantum mechanics experimentally validated in its predictions, but its underlying physical interpretation remains fundamentally unresolved. While the Chinese experiment successfully confirmed the mathematical accuracy of Feynman’s path integral formulas across physical grid points, Hossenfelder concludes that it provides no evidence regarding what particles do when unmonitored, leaving foundational quantum interpretations as unresolved as before.

Source quotes

[0:00] Sabine Hossenfelder: credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch. [0:00] Sabine Hossenfelder: more than mere mathematical artifacts [0:00] Sabine Hossenfelder: reflect a physical aspect of quantum reality. [3:15] New Scientist: Richard Feynman’s 80-year-old quantum postulate has now been validated. [3:15] Sabine Hossenfelder: 9 out of 10 on the bullshit meter [4:33] Sabine Hossenfelder: Feynman is still right, we still don’t know how to interpret quantum physics, and John Bell is still dead.

Original transcript

[0:00] Quantum particles can take every possible path at the same time. This is an 80 years [0:06] old interpretation of mathematics which comes from Richard Feynman. But now a team [0:11] of physicists in China says they have shown that this is more than an interpretation. [0:16] They say they have experimentally proved that it is physical reality. So can we finally stop [0:23] arguing about the interpretation of quantum physics? That’d be huge progress. I mean, [0:28] it’s only been 80 years. By physics standards, that’s basically express delivery. Let’s have a [0:33] look. Stay around until the end for a special offer that will bring your coding skills to [0:39] an entirely new level with boot.dev. In ordinary physics, if you throw a ball, [0:45] then the ball takes one particular path, depending on its initial place and velocity. That’s if you [0:51] throw the ball. If I throw the ball, I’ll tear a body part I didn’t know exists and then the ball [0:56] will get lost in the bushes. Still, physicists insist, it’s all strictly deterministic. [1:02] In quantum physics, this isn’t the case. Rather, Feynman said, [1:06] what you do for a quantum particle is that you look at every possible path. Not just the [1:11] straight connection from a start to an end point, but every connection. For each path [1:16] you calculate a quantity called the amplitude. You sum them all up and from that you get the [1:21] probability of the particle to go to a particular place, but just where it goes is not determined. [1:28] It’s called a “path integral” and Feynman showed that it gives the same result as using [1:34] the wave-function and the Schrödinger equation. But the neat thing about Feynman’s method is [1:40] that it gives you a visual interpretation of what is going on and also tells you exactly [1:45] what is so strange about quantum physics: the particles can go all these different paths at [1:52] the same time. Or do they? This condenses eighty years of quantum foundations into three words. [1:59] A lot of physicists think that yes that is really what the particles do, and not just this, [2:04] many of them even think it’s evidence that we live in a multiverse. For example, [2:09] a press release from Google about their quantum computer claimed they’d done an experiment that, [2:14] quote, gives “credence to the notion that quantum computation occurs in many parallel universes, in [2:21] line with the idea that we live in a multiverse, a prediction first made by David Deutsch.” [2:27] The problem is that by construction of the path integral, you never observe the paths. You only [2:33] observe the particle at the final point. If you were to observe the particle in between, [2:38] then you’d only ever observe one path. The authors of the new paper now say they have shown that, [2:44] yes, the particles actually go all these paths. In their words, the paths are quote: [2:50] “more than mere mathematical artifacts” and “reflect a physical aspect of quantum reality”. [2:56] Let me be clear, the question is not whether Feynman’s path integral is correct or not. [3:02] It makes probably the same predictions as Schrödinger’s formalism. It’s been [3:07] tested a million times and we know it works. The question is whether [3:11] we can experimentally settle what really happens. [3:15] So what did they actually do in their experiment? They used single photons, [3:20] one at a time They let the photons travel from one place to another one and put in a grid of [3:25] measurement locations in between basically, 17 along the direction of travel and 5 transverse [3:31] to it. They take two of these grid points and measure what the amplitude of the photon does [3:36] between them. And that, they find, fits exactly to Feynman’s prediction. So, New Scientist writes: [3:42] “Richard Feynman’s 80-year-old quantum postulate has now been validated.” [3:47] Ok, but in this case the New Scientist article is more reasonable than the paper [3:53] itself. Does it “validate” Feynman’s path integral. Yes, totally. It’s a [3:58] beautiful direct confirmation that the maths really adds up the way he said. Then again, [4:03] no one doubted this in the first place. Does it prove that the particles “really” go all [4:08] possible paths when no one looks? No, of course not. Because they didn’t find out [4:13] what the particles do when you don’t measure them, they put in additional measurements just [4:19] to then find that the particles were on one particular path, which, again no one doubted. [4:25] So, I am sorry, but I give this paper a 9 out of 10 on the bullshit meter. Nice experiment, [4:31] alright, but crappy interpretation. [4:33] Why am I telling you about this anyway? It’s because this is a neat [4:37] illustration of why the foundations of physics are stuck. The problem that these guys try [4:43] to answer is the exact problem that Albert Einstein tried to answer already and failed, [4:48] and which John Bell then tried to answer. Namely: does a quantum particle “really” do [4:54] all the things at the same time when we don’t look, or does it “really” only do one thing. [5:00] So not only did the authors of the paper misunderstand the task, [5:05] they also didn’t realize that this is exactly what Bell’s inequality is about. [5:11] Now as I’ve said many times, the conclusion from Bell only follows once you postulate [5:16] that superdeterminism is ruled out, which has never been ruled out, so actually the current [5:21] situation is still inconclusive. But experiments like the one in the new paper won’t settle it. [5:28] In summary, Feynman is still right, we still don’t know how to interpret [5:33] quantum physics, and John Bell is still dead. The last time I wrote code, it was Fortran this [5:41] is why I decided I need to urgently upgrade my coding skills. If you're in a similar situation, [5:47] I have a recommendation for you. Boot.dev, who've been sponsoring this video. Boot.dev is focused [5:54] on back-end development. So, Python, SQL, Go, the kind of stuff that actually runs the world behind [5:59] the scenes. And you don't just watch lessons, you build real projects, you write code directly in [6:05] the browser, you solve challenges, and you level up like it's a slightly nerdy role-playing game. [6:11] They also have what they call the training grounds. That's basically infinite practice [6:15] problem. I've been meaning for a while to set up a website for all my content, but I want [6:21] to know what's going on with the data and what vulnerabilities to watch out for. And boot.dev has [6:27] been really eyeopening. You can create an account for free, read and watch all the content and try [6:34] a demo of the interactive features. If you decide to go for the full experience, use my code SABINE [6:42] to get 25% off your first year. So go and check this out. Thanks for watching. See you tomorrow.