×
TechCrunch Mobility: ‘Physical AI’ enters the hype machine | TechCrunch

TechCrunch Mobility: ‘Physical AI’ enters the hype machine | TechCrunch

Welcome back to TechCrunch Mobility, your hub for all things “future of transportation.” To get this in your inbox, sign up here for free — just click TechCrunch Mobility!

It’s been a minute, folks! As you might recall, the newsletter took a little holiday break. We’re back and well into 2026. And a lot has happened since the last edition. 

I spent the first week of the year at the Consumer Electronics Show in Las Vegas. And while I wrote about this last January, it’s worth repeating: U.S. automakers have left the building. 

What has filled the void in the Las Vegas Convention Center? Autonomous vehicle tech companies (Zoox, Tensor Auto, Tier IV, and Waymo, which rebranded its Zeekr RT, to name a few), Chinese automakers like Geely and GWM, software and automotive chip companies, and loads of what Nvidia CEO Jensen Huang calls “physical AI.” 

The term, which is sometimes called “embodied AI,” describes the use of AI outside the digital world and into the real, physics-based one. AI models, combined with sensors, cameras, and the motorized controls, allow that physical thing — humanoid robot, drone, autonomous forklift, robotaxi — to detect and understand what’s in this real environment and make decisions to operate within it. And it was all over the place from agriculture and robotics to autonomous vehicles and drones, industrial manufacturing, and wearables. 

Hyundai had one of the busiest and largest exhibits with a near-constant line wrapped around the entrance. The Korean automaker wasn’t showing cars. Nope, it was robots of various forms, including the Atlas humanoid robot, courtesy of its subsidiary Boston Dynamics. There were also innovations that have come out of Hyundai Motor Group Robotics LAB, including a robot that charges electric autonomous vehicles, and a four-wheel electric platform called the Mobile Eccentric Droid (MobEd) that is going into production this year. It seems everyone was embracing and showcasing robotics, particularly humanoids. 

The hype around humanoids, specifically, and physical AI, in general, was palpable. I asked Mobileye co-founder and president Amnon Shashua about this because his company just bought his humanoid robotics startup for $900 million: “What do you say when people tell you humanoid robots are all hype?” 

Techcrunch event

San Francisco
|
October 13-15, 2026

“The internet was also a hype, remember in 2000, the crisis of the internet,” Shashua said. “It did not mean that [the] internet is not a real thing. Hype means that companies are overvalued for a certain period of time, and then they crash. It does not mean that the domain is not real. I believe that the domain of humanoids is real.”

A few notable stories from CES:
Nvidia launches Alpamayo, open AI models that allow autonomous vehicles to ‘think like a human’

This is Uber’s new robotaxi from Lucid and Nuro

Mobileye acquires humanoid robot startup Mentee Robotics for $900M

Now onto the other non-CES and more recent news … 

A little bird

Image Credits:Bryce Durbin

President Trump made comments this week at a Detroit Economic Club meeting about welcoming Chinese automakers into the United States that did not sit well with many in the auto industry, according to insiders I have spoken to. Specifically, I have been told the Alliance for Automotive Innovation (the industry lobbying group) is “freaking out,” one DC insider told me. 

“If they want to come in and build a plant and hire you and hire your friends and your neighbors, that’s great, I love that,” Trump said, according to reporters in attendance. “Let China come in, let Japan come in.”

A couple of notes. Japanese companies like Toyota are already very much in the United States. The bigger hurdle, beyond protests from within the boardrooms of U.S. automakers, is existing law. In 2025, the U.S. Department of Commerce’s Bureau of Industry and Security issued a rule that restricts the import and sale of certain connected vehicles and related hardware and software linked to China or Russia. This essentially bans the sale of Chinese vehicles in the country. 

Avery Ash, who is CEO of SAFE, a nonpartisan organization focused on securing U.S. energy, critical materials, and supply chains, weighed in about the dangers of allowing Chinese automakers to sell their vehicles in the United States. Side note: Ash was on my podcast, the Autonocast, which touches on some of this subject.

“Welcoming Chinese automakers to build cars here in the U.S. will reverse these hard-won accomplishments and put Americans at risk,” he said. ”We’ve seen this strategy backfire in Europe and elsewhere — it would have potentially catastrophic impacts on our automotive industry, have ripple effects on our entire defense industrial base, and make every American less secure.”

Meanwhile, Canada is opening the door to Chinese automakers. Canadian prime minister Mark Carney announced his country will slash its 100% import tax on Chinese EVs to just 6.1%, Sean O’Kane reports.

Got a tip for us to share in the Little Bird section? Email Kirsten Korosec at kirsten.korosec@techcrunch.com or my Signal at kkorosec.07, or email Sean O’Kane at sean.okane@techcrunch.com

Deals!

money the station
Image Credits:Bryce Durbin

Budget carrier Allegiant agreed to buy rival Sun Country Airlines for about $1.5 billion in cash and stock.

Dealerware, which sells software services to automotive OEMs and retailers, was acquired by a group of investors led by Wavecrest Growth Partners and Radian Capital. Automotive Ventures and automotive industry executives David Metter and Devin Daly also participated. The terms were not disclosed.

Long-distance bus and train provider Flix acquired the majority share of European airport transfer-platform Flibco. Luxembourg company SLG will retain some ownership stake in Flibco. Terms weren’t disclosed. 

JetZero, the Long Beach, California, startup developing a midsized triangular aircraft designed to save on fuel, raised $175 million in a Series B round led by B Capital, Bloomberg reported.

Joby Aviation, a company developing electric air taxis, reached an agreement to buy a 700,000-square-foot manufacturing facility in Dayton, Ohio, to support its plans to double production to four aircraft per month in 2027.

Luminar has reached a deal to sell its lidar business to a company called Quantum Computing Inc. for just $22 million. If that seems low, you’re right. Luminar’s valuation peaked in 2021 at $11 billion.

Notable reads and other tidbits

Image Credits:Bryce Durbin

Bluspark Global, a New York-based shipping and supply chain software company, didn’t realize its platform was vulnerable and open to anyone on the internet. Here’s how a security researcher (and TechCrunch) got it fixed.

The Federal Trade Commission finalized an order that bans General Motors and its OnStar telematics service from sharing certain consumer data with consumer reporting agencies. Read the full story on what that means.

InDrive, the company that started as a ride-hailing platform that lets users set the price, is diversifying and starting to execute on its “super app” strategy. That means more in-app advertising across its top 20 markets and expanding grocery delivery to Pakistan. Read the full story here. 

Motional, the majority Hyundai-owned autonomous vehicle company, has rebooted. When Motional paused its operations last year, I wasn’t sure it was going to survive. Other AV companies with big backers have seen their funding disappear in a blink, so it was certainly plausible. But the company is here and with a new AI-first approach. Before you roll your eyes at that term, take a read of my article, which includes a demo ride and an interview with CEO Laura Major. Then feel free to hit my inbox with your thoughts. 

New York governor Kathy Hochul plans to introduce legislation that would effectively legalize robotaxis in the state with the exception of New York City. No details on this yet; I’ve been told it will all be revealed in her executive budget proposal next week. What we do know is the proposal is designed to expand the state’s existing AV pilot program to allow for “the limited deployment of commercial for-hire autonomous passenger vehicles outside New York City.” My article delves deeper into what she shared and gives an update on Waymo’s NYC permit. 

Tesla is ditching the one-time fee option for its Full Self-driving (Supervised) software and will now sell access to the feature through a monthly subscription.

On-demand drone delivery company Wing is bringing its service to another 150 Walmart stores as part of an expanded partnership with the retailer.

Source link
#TechCrunch #Mobility #Physical #enters #hype #machine #TechCrunch

Previously, DoorDash worked with Alphabet’s drone delivery division Wing on a number of pilots in places like Dallas-Fort Worth, Virginia, and Australia. Now the company says it will deploy its own drones, designed specifically for food delivery. The drones will be designed and assembled within the company’s DoorDash Labs division, which houses its robotics and AI teams responsible for building its Dot sidewalk delivery robots.

“We’re designing an aircraft purpose-built for the gaps we see in local commerce, and it complements the short- and long-range designs our partners are scaling,” the company said in a press release. “Our drone is also proudly American-designed and built, with the majority of components made right here in the U.S.” (To be sure, the vast majority of drones are manufactured in China.)

In addition, DoorDash says it will also build all the necessary ground-level infrastructure to enable restaurants to hand off their deliveries to drones. As seen with other drone delivery pilots, there is a complex system of structures and additional equipment that needs to be built on the ground to enable drones to pickup their payloads before embarking on their deliveries. And there are flight plans, airspace management, and drop-off logistics that needs to be established to track the drones in flight. DoorDash says its committed to building all that infrastructure itself.

“The aircraft is what everyone sees,” the company says. “The harder problem is the infrastructure and integration on the ground, and that’s our advantage. Solving hard physical-world problems for local businesses is our focus, from real-time inventory reconciliation to universal handoff systems for drive-throughs, rooftops, or merchant back doors. We are building that complete end-to-end system.”

DoorDash says drone delivery could solve the problem of midrange deliveries that still end up taking longer because couriers are hard to find. The company says more than 20 percent of its orders involve distances of three to five miles, but that those orders can take on average nearly 25 percent longer than shorter deliveries, largely because finding a courier takes longer for these mid-range trips.

“Routing mid-range deliveries to drones lets Dashers focus on the orders they often favor: shorter deliveries that can be done quickly, staying near a high concentration of merchants for optimal routing, and maximizing their earning potential by getting them to their next order faster,” the company says.

DoorDash didn’t reveal which markets it was targeting for its initial launch, noting that it would announce more details in the months to come.

#DoorDash #airborne #drone #delivery #divisionDrones,Food,News,Science,Tech,Transportation">DoorDash is going airborne with new drone delivery divisionDoorDash is launching a new drone delivery program called DoorDash Air. The largest food delivery app in the US said that it has approval from the Federal Aviation Administration that clears the way for drone delivery in the near future.DoorDash said it has received a Part 135 air carrier certification from the FAA that enable it to delivery small packages via drone. Other companies to receive this designation include Alphabet’s Wing, Zipline, and Amazon.Previously, DoorDash worked with Alphabet’s drone delivery division Wing on a number of pilots in places like Dallas-Fort Worth, Virginia, and Australia. Now the company says it will deploy its own drones, designed specifically for food delivery. The drones will be designed and assembled within the company’s DoorDash Labs division, which houses its robotics and AI teams responsible for building its Dot sidewalk delivery robots.“We’re designing an aircraft purpose-built for the gaps we see in local commerce, and it complements the short- and long-range designs our partners are scaling,” the company said in a press release. “Our drone is also proudly American-designed and built, with the majority of components made right here in the U.S.” (To be sure, the vast majority of drones are manufactured in China.)In addition, DoorDash says it will also build all the necessary ground-level infrastructure to enable restaurants to hand off their deliveries to drones. As seen with other drone delivery pilots, there is a complex system of structures and additional equipment that needs to be built on the ground to enable drones to pickup their payloads before embarking on their deliveries. And there are flight plans, airspace management, and drop-off logistics that needs to be established to track the drones in flight. DoorDash says its committed to building all that infrastructure itself.“The aircraft is what everyone sees,” the company says. “The harder problem is the infrastructure and integration on the ground, and that’s our advantage. Solving hard physical-world problems for local businesses is our focus, from real-time inventory reconciliation to universal handoff systems for drive-throughs, rooftops, or merchant back doors. We are building that complete end-to-end system.”DoorDash says drone delivery could solve the problem of midrange deliveries that still end up taking longer because couriers are hard to find. The company says more than 20 percent of its orders involve distances of three to five miles, but that those orders can take on average nearly 25 percent longer than shorter deliveries, largely because finding a courier takes longer for these mid-range trips.“Routing mid-range deliveries to drones lets Dashers focus on the orders they often favor: shorter deliveries that can be done quickly, staying near a high concentration of merchants for optimal routing, and maximizing their earning potential by getting them to their next order faster,” the company says.DoorDash didn’t reveal which markets it was targeting for its initial launch, noting that it would announce more details in the months to come.#DoorDash #airborne #drone #delivery #divisionDrones,Food,News,Science,Tech,Transportation

Previously, DoorDash worked with Alphabet’s drone delivery division Wing on a number of pilots in places like Dallas-Fort Worth, Virginia, and Australia. Now the company says it will deploy its own drones, designed specifically for food delivery. The drones will be designed and assembled within the company’s DoorDash Labs division, which houses its robotics and AI teams responsible for building its Dot sidewalk delivery robots.

“We’re designing an aircraft purpose-built for the gaps we see in local commerce, and it complements the short- and long-range designs our partners are scaling,” the company said in a press release. “Our drone is also proudly American-designed and built, with the majority of components made right here in the U.S.” (To be sure, the vast majority of drones are manufactured in China.)

In addition, DoorDash says it will also build all the necessary ground-level infrastructure to enable restaurants to hand off their deliveries to drones. As seen with other drone delivery pilots, there is a complex system of structures and additional equipment that needs to be built on the ground to enable drones to pickup their payloads before embarking on their deliveries. And there are flight plans, airspace management, and drop-off logistics that needs to be established to track the drones in flight. DoorDash says its committed to building all that infrastructure itself.

“The aircraft is what everyone sees,” the company says. “The harder problem is the infrastructure and integration on the ground, and that’s our advantage. Solving hard physical-world problems for local businesses is our focus, from real-time inventory reconciliation to universal handoff systems for drive-throughs, rooftops, or merchant back doors. We are building that complete end-to-end system.”

DoorDash says drone delivery could solve the problem of midrange deliveries that still end up taking longer because couriers are hard to find. The company says more than 20 percent of its orders involve distances of three to five miles, but that those orders can take on average nearly 25 percent longer than shorter deliveries, largely because finding a courier takes longer for these mid-range trips.

“Routing mid-range deliveries to drones lets Dashers focus on the orders they often favor: shorter deliveries that can be done quickly, staying near a high concentration of merchants for optimal routing, and maximizing their earning potential by getting them to their next order faster,” the company says.

DoorDash didn’t reveal which markets it was targeting for its initial launch, noting that it would announce more details in the months to come.

#DoorDash #airborne #drone #delivery #divisionDrones,Food,News,Science,Tech,Transportation">DoorDash is going airborne with new drone delivery division

DoorDash is launching a new drone delivery program called DoorDash Air. The largest food delivery app in the US said that it has approval from the Federal Aviation Administration that clears the way for drone delivery in the near future.

DoorDash said it has received a Part 135 air carrier certification from the FAA that enable it to delivery small packages via drone. Other companies to receive this designation include Alphabet’s Wing, Zipline, and Amazon.

Previously, DoorDash worked with Alphabet’s drone delivery division Wing on a number of pilots in places like Dallas-Fort Worth, Virginia, and Australia. Now the company says it will deploy its own drones, designed specifically for food delivery. The drones will be designed and assembled within the company’s DoorDash Labs division, which houses its robotics and AI teams responsible for building its Dot sidewalk delivery robots.

“We’re designing an aircraft purpose-built for the gaps we see in local commerce, and it complements the short- and long-range designs our partners are scaling,” the company said in a press release. “Our drone is also proudly American-designed and built, with the majority of components made right here in the U.S.” (To be sure, the vast majority of drones are manufactured in China.)

In addition, DoorDash says it will also build all the necessary ground-level infrastructure to enable restaurants to hand off their deliveries to drones. As seen with other drone delivery pilots, there is a complex system of structures and additional equipment that needs to be built on the ground to enable drones to pickup their payloads before embarking on their deliveries. And there are flight plans, airspace management, and drop-off logistics that needs to be established to track the drones in flight. DoorDash says its committed to building all that infrastructure itself.

“The aircraft is what everyone sees,” the company says. “The harder problem is the infrastructure and integration on the ground, and that’s our advantage. Solving hard physical-world problems for local businesses is our focus, from real-time inventory reconciliation to universal handoff systems for drive-throughs, rooftops, or merchant back doors. We are building that complete end-to-end system.”

DoorDash says drone delivery could solve the problem of midrange deliveries that still end up taking longer because couriers are hard to find. The company says more than 20 percent of its orders involve distances of three to five miles, but that those orders can take on average nearly 25 percent longer than shorter deliveries, largely because finding a courier takes longer for these mid-range trips.

“Routing mid-range deliveries to drones lets Dashers focus on the orders they often favor: shorter deliveries that can be done quickly, staying near a high concentration of merchants for optimal routing, and maximizing their earning potential by getting them to their next order faster,” the company says.

DoorDash didn’t reveal which markets it was targeting for its initial launch, noting that it would announce more details in the months to come.

#DoorDash #airborne #drone #delivery #divisionDrones,Food,News,Science,Tech,Transportation
New York-based AI detection startup Pangram is on a mission to combat the AI slop infestation spreading across the internet, and it just raised $9 million on a bet that demand for tools that distinguish human-generated content from AI-generated text will only grow. 

Pangram’s fundraise — led by Menlo Ventures, with participation from Haystack, ScOp, Script Capital, and Cadenza — comes as the startup also launches its next-generation AI text detection model, Pangram 4, and an AI image detection model, Pangram Image. 

Pangram says the new text detection model is over 99% accurate at finding AI-assisted writing and mixed human-AI content, plus it can more easily detect AI humanizer programs. The AI image detector is only available via research preview for now; Pangram plans to release it more widely in the coming weeks.

Stanford AI and machine learning grads Max Spero and Bradley Emi launched Pangram about two years ago, after the launch of ChatGPT opened the floodgates for an internet full of bots, AI-generated SEO slop content, and what Spero calls “LLM-powered Russian disinformation campaigns and UAE-influenced campaigns on Twitter.”

“I think it’s just incredibly valuable to know whether what you’re looking at is something that’s AI-generated or not,” Spero told TechCrunch. “Especially text that you’re reading, because it changes how people approach the text. Is this something that I’m going to have to look out for hallucinations and jump in skeptically, or is this something that I trust was well-researched from an actual journalist?”

Pangram’s AI detection system is essentially a large machine learning model that was trained on tens of millions of known human documents. The startup then created a “synthetic mirror” for each document, replicating the topic, length, and tone of voice, but written by a frontier LLM. 

“Our model is learning the stylistic differences and the choices that AI makes consistently and is able to use that to learn what makes something AI-generated with high confidence,” Spero said, adding that the AI detector isn’t relying on copy-paste metadata or hidden watermarks. 

For Pangram, AI detection isn’t just about whether or not a piece of text was written entirely by AI. It’s also about distinguishing between levels of AI assistance — like in the case of someone who writes something themselves, but then asks AI to edit or clean it up. Spero believes AI assistance can be acceptable, just so long as the writer discloses their use of AI. 

As AI content floods the internet, Pangram raises M to detect it | TechCrunch
New York-based AI detection startup Pangram is on a mission to combat the AI slop infestation spreading across the internet, and it just raised  million on a bet that demand for tools that distinguish human-generated content from AI-generated text will only grow. 

Pangram’s fundraise — led by Menlo Ventures, with participation from Haystack, ScOp, Script Capital, and Cadenza — comes as the startup also launches its next-generation AI text detection model, Pangram 4, and an AI image detection model, Pangram Image. 







Pangram says the new text detection model is over 99% accurate at finding AI-assisted writing and mixed human-AI content, plus it can more easily detect AI humanizer programs. The AI image detector is only available via research preview for now; Pangram plans to release it more widely in the coming weeks.

Stanford AI and machine learning grads Max Spero and Bradley Emi launched Pangram about two years ago, after the launch of ChatGPT opened the floodgates for an internet full of bots, AI-generated SEO slop content, and what Spero calls “LLM-powered Russian disinformation campaigns and UAE-influenced campaigns on Twitter.”

“I think it’s just incredibly valuable to know whether what you’re looking at is something that’s AI-generated or not,” Spero told TechCrunch. “Especially text that you’re reading, because it changes how people approach the text. Is this something that I’m going to have to look out for hallucinations and jump in skeptically, or is this something that I trust was well-researched from an actual journalist?”

Pangram’s AI detection system is essentially a large machine learning model that was trained on tens of millions of known human documents. The startup then created a “synthetic mirror” for each document, replicating the topic, length, and tone of voice, but written by a frontier LLM. 

“Our model is learning the stylistic differences and the choices that AI makes consistently and is able to use that to learn what makes something AI-generated with high confidence,” Spero said, adding that the AI detector isn’t relying on copy-paste metadata or hidden watermarks. 


For Pangram, AI detection isn’t just about whether or not a piece of text was written entirely by AI. It’s also about distinguishing between levels of AI assistance — like in the case of someone who writes something themselves, but then asks AI to edit or clean it up. Spero believes AI assistance can be acceptable, just so long as the writer discloses their use of AI. 

Image Credits:Pangram

Pangram’s emergence comes at a time when AI usage is becoming more commonplace. In some cases, like the Canadian politician who read an AI prompt aloud in a speech to lawmakers, the mistakes result in ridicule. In other cases, as with certain lawyers making their case using fake citations created by ChatGPT, the consequences could be sanctions and fines.  

That backlash isn’t just costing individuals embarrassment or sanctions — it’s starting to show up in institutional rules, too.







The open-access archive arXiv introduced a new enforcement policy this year, stating that submissions containing evidence that authors failed to review LLM output (like hallucinated references or meta comments such as, “Would you like me to make any changes?”) can trigger a one-year submission ban.

Pangram isn’t the only one betting that AI detection will become more sought after. Competitors like Winston AI, Originality.ai, Copyleaks, and GPTZero are are chasing the same demand, each building its own detector.

Pangram’s technology, while not perfect, could help fuel the resistance to accepting the AI-generated content flooding the internet, the courtroom, and academic papers. 

Users can access Pangram via a -per-month subscription on the web or download the Chrome extension, which automatically labels posts in real time on X, LinkedIn, Substack, Reddit, and Medium. It also provides a feed health score with a percentage breakdown of human versus AI content on your screen. 

Pangram also offers its technology via API. Notably, Substack recently integrated Pangram’s technology into its platform to show readers which of their favorite authors write their newsletters using AI. Other API customers include Quora, schools and universities, publishers and agents, and recruiters, among others, per Spero. 

Does Pangram work?

Pangram detected AI-generated content even when lightly edited by a human. Image Credits:Pangram/TechCrunch

Spero said roughly one in 10,000 human documents are incorrectly labeled as AI with Pangram’s model, so I decided to put it to the test. The text detection model was very impressive but not perfect. It easily flagged entirely AI-generated news articles written by both ChatGPT and Claude, and was rarely fooled by my attempts to edit the AI-generated text into sounding more human. At the same time, Pangram did flag sentences that I completely rewrote as being AI-written. Pangram also wasn’t at all fooled by my attempts to prompt ChatGPT and Claude into evading AI detectors when generating content.  

I also gave ChatGPT and Claude one of my own articles and asked them to polish it up. Pangram gave it a 13% AI assisted score, which was probably close to accurate, but the model was able to detect subtle word-choice changes in some sentences and ignored them in others. It also flagged some sentences as AI-assisted when they were human written. That was notable because when I gave Pangram that same article in its entirety, as I had written it, it got a 100% human score. 

Maybe the problem was that news articles can be a bit dry and could easily sound like AI. So I tried a different tactic. I tested Pangram on my own more voicey, personal Substack newsletter content, pasting the first half of the text into Pangram and then asking ChatGPT and Claude to copy my style and write the second half. For the most part, Pangram easily detected human-written text versus AI-written text.







My limited testing of Pangram’s new image detection model turned out to be equally impressive. 

Image Credits:Pangram/TechCrunch

Pangram’s AI image detection system promises to spot AI-generated images across AI models, unlike OpenAI’s or Google DeepMind’s watermark-based checks, which mostly detect their own output. It works on pixel-level distributions, learning subtle statistical differences between real photos and AI-generated images. Spero says the model can even detect an AI image that appears inside a real-world photo. 

In my testing, the model easily detected AI-generated imagery, whether it was photorealistic or cartoonish. I can also confirm the model could detect an AI image appearing in a real-world photo — the heat map Pangram provides clearly lighting up over the image — though in one instance it incorrectly labeled a photo of an AI-generated image as human content. 

Spero says he doesn’t want his technology to fuel a witch hunt against people using AI for writing, but that there needs to be some sort of mechanism to push back against the slop. 

“The future that I see is that AI content just continues to proliferate,” Spero said. “We’re getting new GPUs faster than new people are being born. If we do not actively discriminate in favor of human content, then we’re just gonna see more and more AI, and it’s just gonna drown out any human signal that we have.”
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.#content #floods #internet #Pangram #raises #detect #TechCrunchai slop
Image Credits:Pangram

Pangram’s emergence comes at a time when AI usage is becoming more commonplace. In some cases, like the Canadian politician who read an AI prompt aloud in a speech to lawmakers, the mistakes result in ridicule. In other cases, as with certain lawyers making their case using fake citations created by ChatGPT, the consequences could be sanctions and fines.  

That backlash isn’t just costing individuals embarrassment or sanctions — it’s starting to show up in institutional rules, too.

The open-access archive arXiv introduced a new enforcement policy this year, stating that submissions containing evidence that authors failed to review LLM output (like hallucinated references or meta comments such as, “Would you like me to make any changes?”) can trigger a one-year submission ban.

Pangram isn’t the only one betting that AI detection will become more sought after. Competitors like Winston AI, Originality.ai, Copyleaks, and GPTZero are are chasing the same demand, each building its own detector.

Pangram’s technology, while not perfect, could help fuel the resistance to accepting the AI-generated content flooding the internet, the courtroom, and academic papers. 

Users can access Pangram via a $20-per-month subscription on the web or download the Chrome extension, which automatically labels posts in real time on X, LinkedIn, Substack, Reddit, and Medium. It also provides a feed health score with a percentage breakdown of human versus AI content on your screen. 

Pangram also offers its technology via API. Notably, Substack recently integrated Pangram’s technology into its platform to show readers which of their favorite authors write their newsletters using AI. Other API customers include Quora, schools and universities, publishers and agents, and recruiters, among others, per Spero. 

Does Pangram work?

Pangram detected AI-generated content even when lightly edited by a human. Image Credits:Pangram/TechCrunch

Spero said roughly one in 10,000 human documents are incorrectly labeled as AI with Pangram’s model, so I decided to put it to the test. The text detection model was very impressive but not perfect. It easily flagged entirely AI-generated news articles written by both ChatGPT and Claude, and was rarely fooled by my attempts to edit the AI-generated text into sounding more human. At the same time, Pangram did flag sentences that I completely rewrote as being AI-written. Pangram also wasn’t at all fooled by my attempts to prompt ChatGPT and Claude into evading AI detectors when generating content.  

I also gave ChatGPT and Claude one of my own articles and asked them to polish it up. Pangram gave it a 13% AI assisted score, which was probably close to accurate, but the model was able to detect subtle word-choice changes in some sentences and ignored them in others. It also flagged some sentences as AI-assisted when they were human written. That was notable because when I gave Pangram that same article in its entirety, as I had written it, it got a 100% human score. 

Maybe the problem was that news articles can be a bit dry and could easily sound like AI. So I tried a different tactic. I tested Pangram on my own more voicey, personal Substack newsletter content, pasting the first half of the text into Pangram and then asking ChatGPT and Claude to copy my style and write the second half. For the most part, Pangram easily detected human-written text versus AI-written text.

My limited testing of Pangram’s new image detection model turned out to be equally impressive. 

Image Credits:Pangram/TechCrunch

Pangram’s AI image detection system promises to spot AI-generated images across AI models, unlike OpenAI’s or Google DeepMind’s watermark-based checks, which mostly detect their own output. It works on pixel-level distributions, learning subtle statistical differences between real photos and AI-generated images. Spero says the model can even detect an AI image that appears inside a real-world photo. 

In my testing, the model easily detected AI-generated imagery, whether it was photorealistic or cartoonish. I can also confirm the model could detect an AI image appearing in a real-world photo — the heat map Pangram provides clearly lighting up over the image — though in one instance it incorrectly labeled a photo of an AI-generated image as human content. 

Spero says he doesn’t want his technology to fuel a witch hunt against people using AI for writing, but that there needs to be some sort of mechanism to push back against the slop. 

“The future that I see is that AI content just continues to proliferate,” Spero said. “We’re getting new GPUs faster than new people are being born. If we do not actively discriminate in favor of human content, then we’re just gonna see more and more AI, and it’s just gonna drown out any human signal that we have.”

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

#content #floods #internet #Pangram #raises #detect #TechCrunchai slop">As AI content floods the internet, Pangram raises M to detect it | TechCrunch
New York-based AI detection startup Pangram is on a mission to combat the AI slop infestation spreading across the internet, and it just raised  million on a bet that demand for tools that distinguish human-generated content from AI-generated text will only grow. 

Pangram’s fundraise — led by Menlo Ventures, with participation from Haystack, ScOp, Script Capital, and Cadenza — comes as the startup also launches its next-generation AI text detection model, Pangram 4, and an AI image detection model, Pangram Image. 







Pangram says the new text detection model is over 99% accurate at finding AI-assisted writing and mixed human-AI content, plus it can more easily detect AI humanizer programs. The AI image detector is only available via research preview for now; Pangram plans to release it more widely in the coming weeks.

Stanford AI and machine learning grads Max Spero and Bradley Emi launched Pangram about two years ago, after the launch of ChatGPT opened the floodgates for an internet full of bots, AI-generated SEO slop content, and what Spero calls “LLM-powered Russian disinformation campaigns and UAE-influenced campaigns on Twitter.”

“I think it’s just incredibly valuable to know whether what you’re looking at is something that’s AI-generated or not,” Spero told TechCrunch. “Especially text that you’re reading, because it changes how people approach the text. Is this something that I’m going to have to look out for hallucinations and jump in skeptically, or is this something that I trust was well-researched from an actual journalist?”

Pangram’s AI detection system is essentially a large machine learning model that was trained on tens of millions of known human documents. The startup then created a “synthetic mirror” for each document, replicating the topic, length, and tone of voice, but written by a frontier LLM. 

“Our model is learning the stylistic differences and the choices that AI makes consistently and is able to use that to learn what makes something AI-generated with high confidence,” Spero said, adding that the AI detector isn’t relying on copy-paste metadata or hidden watermarks. 


For Pangram, AI detection isn’t just about whether or not a piece of text was written entirely by AI. It’s also about distinguishing between levels of AI assistance — like in the case of someone who writes something themselves, but then asks AI to edit or clean it up. Spero believes AI assistance can be acceptable, just so long as the writer discloses their use of AI. 

Image Credits:Pangram

Pangram’s emergence comes at a time when AI usage is becoming more commonplace. In some cases, like the Canadian politician who read an AI prompt aloud in a speech to lawmakers, the mistakes result in ridicule. In other cases, as with certain lawyers making their case using fake citations created by ChatGPT, the consequences could be sanctions and fines.  

That backlash isn’t just costing individuals embarrassment or sanctions — it’s starting to show up in institutional rules, too.







The open-access archive arXiv introduced a new enforcement policy this year, stating that submissions containing evidence that authors failed to review LLM output (like hallucinated references or meta comments such as, “Would you like me to make any changes?”) can trigger a one-year submission ban.

Pangram isn’t the only one betting that AI detection will become more sought after. Competitors like Winston AI, Originality.ai, Copyleaks, and GPTZero are are chasing the same demand, each building its own detector.

Pangram’s technology, while not perfect, could help fuel the resistance to accepting the AI-generated content flooding the internet, the courtroom, and academic papers. 

Users can access Pangram via a -per-month subscription on the web or download the Chrome extension, which automatically labels posts in real time on X, LinkedIn, Substack, Reddit, and Medium. It also provides a feed health score with a percentage breakdown of human versus AI content on your screen. 

Pangram also offers its technology via API. Notably, Substack recently integrated Pangram’s technology into its platform to show readers which of their favorite authors write their newsletters using AI. Other API customers include Quora, schools and universities, publishers and agents, and recruiters, among others, per Spero. 

Does Pangram work?

Pangram detected AI-generated content even when lightly edited by a human. Image Credits:Pangram/TechCrunch

Spero said roughly one in 10,000 human documents are incorrectly labeled as AI with Pangram’s model, so I decided to put it to the test. The text detection model was very impressive but not perfect. It easily flagged entirely AI-generated news articles written by both ChatGPT and Claude, and was rarely fooled by my attempts to edit the AI-generated text into sounding more human. At the same time, Pangram did flag sentences that I completely rewrote as being AI-written. Pangram also wasn’t at all fooled by my attempts to prompt ChatGPT and Claude into evading AI detectors when generating content.  

I also gave ChatGPT and Claude one of my own articles and asked them to polish it up. Pangram gave it a 13% AI assisted score, which was probably close to accurate, but the model was able to detect subtle word-choice changes in some sentences and ignored them in others. It also flagged some sentences as AI-assisted when they were human written. That was notable because when I gave Pangram that same article in its entirety, as I had written it, it got a 100% human score. 

Maybe the problem was that news articles can be a bit dry and could easily sound like AI. So I tried a different tactic. I tested Pangram on my own more voicey, personal Substack newsletter content, pasting the first half of the text into Pangram and then asking ChatGPT and Claude to copy my style and write the second half. For the most part, Pangram easily detected human-written text versus AI-written text.







My limited testing of Pangram’s new image detection model turned out to be equally impressive. 

Image Credits:Pangram/TechCrunch

Pangram’s AI image detection system promises to spot AI-generated images across AI models, unlike OpenAI’s or Google DeepMind’s watermark-based checks, which mostly detect their own output. It works on pixel-level distributions, learning subtle statistical differences between real photos and AI-generated images. Spero says the model can even detect an AI image that appears inside a real-world photo. 

In my testing, the model easily detected AI-generated imagery, whether it was photorealistic or cartoonish. I can also confirm the model could detect an AI image appearing in a real-world photo — the heat map Pangram provides clearly lighting up over the image — though in one instance it incorrectly labeled a photo of an AI-generated image as human content. 

Spero says he doesn’t want his technology to fuel a witch hunt against people using AI for writing, but that there needs to be some sort of mechanism to push back against the slop. 

“The future that I see is that AI content just continues to proliferate,” Spero said. “We’re getting new GPUs faster than new people are being born. If we do not actively discriminate in favor of human content, then we’re just gonna see more and more AI, and it’s just gonna drown out any human signal that we have.”
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.#content #floods #internet #Pangram #raises #detect #TechCrunchai slop

Pangram is on a mission to combat the AI slop infestation spreading across the internet, and it just raised $9 million on a bet that demand for tools that distinguish human-generated content from AI-generated text will only grow. 

Pangram’s fundraise — led by Menlo Ventures, with participation from Haystack, ScOp, Script Capital, and Cadenza — comes as the startup also launches its next-generation AI text detection model, Pangram 4, and an AI image detection model, Pangram Image. 

Pangram says the new text detection model is over 99% accurate at finding AI-assisted writing and mixed human-AI content, plus it can more easily detect AI humanizer programs. The AI image detector is only available via research preview for now; Pangram plans to release it more widely in the coming weeks.

Stanford AI and machine learning grads Max Spero and Bradley Emi launched Pangram about two years ago, after the launch of ChatGPT opened the floodgates for an internet full of bots, AI-generated SEO slop content, and what Spero calls “LLM-powered Russian disinformation campaigns and UAE-influenced campaigns on Twitter.”

“I think it’s just incredibly valuable to know whether what you’re looking at is something that’s AI-generated or not,” Spero told TechCrunch. “Especially text that you’re reading, because it changes how people approach the text. Is this something that I’m going to have to look out for hallucinations and jump in skeptically, or is this something that I trust was well-researched from an actual journalist?”

Pangram’s AI detection system is essentially a large machine learning model that was trained on tens of millions of known human documents. The startup then created a “synthetic mirror” for each document, replicating the topic, length, and tone of voice, but written by a frontier LLM. 

“Our model is learning the stylistic differences and the choices that AI makes consistently and is able to use that to learn what makes something AI-generated with high confidence,” Spero said, adding that the AI detector isn’t relying on copy-paste metadata or hidden watermarks. 

For Pangram, AI detection isn’t just about whether or not a piece of text was written entirely by AI. It’s also about distinguishing between levels of AI assistance — like in the case of someone who writes something themselves, but then asks AI to edit or clean it up. Spero believes AI assistance can be acceptable, just so long as the writer discloses their use of AI. 

As AI content floods the internet, Pangram raises M to detect it | TechCrunch
New York-based AI detection startup Pangram is on a mission to combat the AI slop infestation spreading across the internet, and it just raised  million on a bet that demand for tools that distinguish human-generated content from AI-generated text will only grow. 

Pangram’s fundraise — led by Menlo Ventures, with participation from Haystack, ScOp, Script Capital, and Cadenza — comes as the startup also launches its next-generation AI text detection model, Pangram 4, and an AI image detection model, Pangram Image. 







Pangram says the new text detection model is over 99% accurate at finding AI-assisted writing and mixed human-AI content, plus it can more easily detect AI humanizer programs. The AI image detector is only available via research preview for now; Pangram plans to release it more widely in the coming weeks.

Stanford AI and machine learning grads Max Spero and Bradley Emi launched Pangram about two years ago, after the launch of ChatGPT opened the floodgates for an internet full of bots, AI-generated SEO slop content, and what Spero calls “LLM-powered Russian disinformation campaigns and UAE-influenced campaigns on Twitter.”

“I think it’s just incredibly valuable to know whether what you’re looking at is something that’s AI-generated or not,” Spero told TechCrunch. “Especially text that you’re reading, because it changes how people approach the text. Is this something that I’m going to have to look out for hallucinations and jump in skeptically, or is this something that I trust was well-researched from an actual journalist?”

Pangram’s AI detection system is essentially a large machine learning model that was trained on tens of millions of known human documents. The startup then created a “synthetic mirror” for each document, replicating the topic, length, and tone of voice, but written by a frontier LLM. 

“Our model is learning the stylistic differences and the choices that AI makes consistently and is able to use that to learn what makes something AI-generated with high confidence,” Spero said, adding that the AI detector isn’t relying on copy-paste metadata or hidden watermarks. 


For Pangram, AI detection isn’t just about whether or not a piece of text was written entirely by AI. It’s also about distinguishing between levels of AI assistance — like in the case of someone who writes something themselves, but then asks AI to edit or clean it up. Spero believes AI assistance can be acceptable, just so long as the writer discloses their use of AI. 

Image Credits:Pangram

Pangram’s emergence comes at a time when AI usage is becoming more commonplace. In some cases, like the Canadian politician who read an AI prompt aloud in a speech to lawmakers, the mistakes result in ridicule. In other cases, as with certain lawyers making their case using fake citations created by ChatGPT, the consequences could be sanctions and fines.  

That backlash isn’t just costing individuals embarrassment or sanctions — it’s starting to show up in institutional rules, too.







The open-access archive arXiv introduced a new enforcement policy this year, stating that submissions containing evidence that authors failed to review LLM output (like hallucinated references or meta comments such as, “Would you like me to make any changes?”) can trigger a one-year submission ban.

Pangram isn’t the only one betting that AI detection will become more sought after. Competitors like Winston AI, Originality.ai, Copyleaks, and GPTZero are are chasing the same demand, each building its own detector.

Pangram’s technology, while not perfect, could help fuel the resistance to accepting the AI-generated content flooding the internet, the courtroom, and academic papers. 

Users can access Pangram via a -per-month subscription on the web or download the Chrome extension, which automatically labels posts in real time on X, LinkedIn, Substack, Reddit, and Medium. It also provides a feed health score with a percentage breakdown of human versus AI content on your screen. 

Pangram also offers its technology via API. Notably, Substack recently integrated Pangram’s technology into its platform to show readers which of their favorite authors write their newsletters using AI. Other API customers include Quora, schools and universities, publishers and agents, and recruiters, among others, per Spero. 

Does Pangram work?

Pangram detected AI-generated content even when lightly edited by a human. Image Credits:Pangram/TechCrunch

Spero said roughly one in 10,000 human documents are incorrectly labeled as AI with Pangram’s model, so I decided to put it to the test. The text detection model was very impressive but not perfect. It easily flagged entirely AI-generated news articles written by both ChatGPT and Claude, and was rarely fooled by my attempts to edit the AI-generated text into sounding more human. At the same time, Pangram did flag sentences that I completely rewrote as being AI-written. Pangram also wasn’t at all fooled by my attempts to prompt ChatGPT and Claude into evading AI detectors when generating content.  

I also gave ChatGPT and Claude one of my own articles and asked them to polish it up. Pangram gave it a 13% AI assisted score, which was probably close to accurate, but the model was able to detect subtle word-choice changes in some sentences and ignored them in others. It also flagged some sentences as AI-assisted when they were human written. That was notable because when I gave Pangram that same article in its entirety, as I had written it, it got a 100% human score. 

Maybe the problem was that news articles can be a bit dry and could easily sound like AI. So I tried a different tactic. I tested Pangram on my own more voicey, personal Substack newsletter content, pasting the first half of the text into Pangram and then asking ChatGPT and Claude to copy my style and write the second half. For the most part, Pangram easily detected human-written text versus AI-written text.







My limited testing of Pangram’s new image detection model turned out to be equally impressive. 

Image Credits:Pangram/TechCrunch

Pangram’s AI image detection system promises to spot AI-generated images across AI models, unlike OpenAI’s or Google DeepMind’s watermark-based checks, which mostly detect their own output. It works on pixel-level distributions, learning subtle statistical differences between real photos and AI-generated images. Spero says the model can even detect an AI image that appears inside a real-world photo. 

In my testing, the model easily detected AI-generated imagery, whether it was photorealistic or cartoonish. I can also confirm the model could detect an AI image appearing in a real-world photo — the heat map Pangram provides clearly lighting up over the image — though in one instance it incorrectly labeled a photo of an AI-generated image as human content. 

Spero says he doesn’t want his technology to fuel a witch hunt against people using AI for writing, but that there needs to be some sort of mechanism to push back against the slop. 

“The future that I see is that AI content just continues to proliferate,” Spero said. “We’re getting new GPUs faster than new people are being born. If we do not actively discriminate in favor of human content, then we’re just gonna see more and more AI, and it’s just gonna drown out any human signal that we have.”
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.#content #floods #internet #Pangram #raises #detect #TechCrunchai slop
Image Credits:Pangram

Pangram’s emergence comes at a time when AI usage is becoming more commonplace. In some cases, like the Canadian politician who read an AI prompt aloud in a speech to lawmakers, the mistakes result in ridicule. In other cases, as with certain lawyers making their case using fake citations created by ChatGPT, the consequences could be sanctions and fines.  

That backlash isn’t just costing individuals embarrassment or sanctions — it’s starting to show up in institutional rules, too.

The open-access archive arXiv introduced a new enforcement policy this year, stating that submissions containing evidence that authors failed to review LLM output (like hallucinated references or meta comments such as, “Would you like me to make any changes?”) can trigger a one-year submission ban.

Pangram isn’t the only one betting that AI detection will become more sought after. Competitors like Winston AI, Originality.ai, Copyleaks, and GPTZero are are chasing the same demand, each building its own detector.

Pangram’s technology, while not perfect, could help fuel the resistance to accepting the AI-generated content flooding the internet, the courtroom, and academic papers. 

Users can access Pangram via a $20-per-month subscription on the web or download the Chrome extension, which automatically labels posts in real time on X, LinkedIn, Substack, Reddit, and Medium. It also provides a feed health score with a percentage breakdown of human versus AI content on your screen. 

Pangram also offers its technology via API. Notably, Substack recently integrated Pangram’s technology into its platform to show readers which of their favorite authors write their newsletters using AI. Other API customers include Quora, schools and universities, publishers and agents, and recruiters, among others, per Spero. 

Does Pangram work?

Pangram detected AI-generated content even when lightly edited by a human. Image Credits:Pangram/TechCrunch

Spero said roughly one in 10,000 human documents are incorrectly labeled as AI with Pangram’s model, so I decided to put it to the test. The text detection model was very impressive but not perfect. It easily flagged entirely AI-generated news articles written by both ChatGPT and Claude, and was rarely fooled by my attempts to edit the AI-generated text into sounding more human. At the same time, Pangram did flag sentences that I completely rewrote as being AI-written. Pangram also wasn’t at all fooled by my attempts to prompt ChatGPT and Claude into evading AI detectors when generating content.  

I also gave ChatGPT and Claude one of my own articles and asked them to polish it up. Pangram gave it a 13% AI assisted score, which was probably close to accurate, but the model was able to detect subtle word-choice changes in some sentences and ignored them in others. It also flagged some sentences as AI-assisted when they were human written. That was notable because when I gave Pangram that same article in its entirety, as I had written it, it got a 100% human score. 

Maybe the problem was that news articles can be a bit dry and could easily sound like AI. So I tried a different tactic. I tested Pangram on my own more voicey, personal Substack newsletter content, pasting the first half of the text into Pangram and then asking ChatGPT and Claude to copy my style and write the second half. For the most part, Pangram easily detected human-written text versus AI-written text.

My limited testing of Pangram’s new image detection model turned out to be equally impressive. 

Image Credits:Pangram/TechCrunch

Pangram’s AI image detection system promises to spot AI-generated images across AI models, unlike OpenAI’s or Google DeepMind’s watermark-based checks, which mostly detect their own output. It works on pixel-level distributions, learning subtle statistical differences between real photos and AI-generated images. Spero says the model can even detect an AI image that appears inside a real-world photo. 

In my testing, the model easily detected AI-generated imagery, whether it was photorealistic or cartoonish. I can also confirm the model could detect an AI image appearing in a real-world photo — the heat map Pangram provides clearly lighting up over the image — though in one instance it incorrectly labeled a photo of an AI-generated image as human content. 

Spero says he doesn’t want his technology to fuel a witch hunt against people using AI for writing, but that there needs to be some sort of mechanism to push back against the slop. 

“The future that I see is that AI content just continues to proliferate,” Spero said. “We’re getting new GPUs faster than new people are being born. If we do not actively discriminate in favor of human content, then we’re just gonna see more and more AI, and it’s just gonna drown out any human signal that we have.”

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

#content #floods #internet #Pangram #raises #detect #TechCrunchai slop">As AI content floods the internet, Pangram raises $9M to detect it | TechCrunch

New York-based AI detection startup Pangram is on a mission to combat the AI slop infestation spreading across the internet, and it just raised $9 million on a bet that demand for tools that distinguish human-generated content from AI-generated text will only grow. 

Pangram’s fundraise — led by Menlo Ventures, with participation from Haystack, ScOp, Script Capital, and Cadenza — comes as the startup also launches its next-generation AI text detection model, Pangram 4, and an AI image detection model, Pangram Image. 

Pangram says the new text detection model is over 99% accurate at finding AI-assisted writing and mixed human-AI content, plus it can more easily detect AI humanizer programs. The AI image detector is only available via research preview for now; Pangram plans to release it more widely in the coming weeks.

Stanford AI and machine learning grads Max Spero and Bradley Emi launched Pangram about two years ago, after the launch of ChatGPT opened the floodgates for an internet full of bots, AI-generated SEO slop content, and what Spero calls “LLM-powered Russian disinformation campaigns and UAE-influenced campaigns on Twitter.”

“I think it’s just incredibly valuable to know whether what you’re looking at is something that’s AI-generated or not,” Spero told TechCrunch. “Especially text that you’re reading, because it changes how people approach the text. Is this something that I’m going to have to look out for hallucinations and jump in skeptically, or is this something that I trust was well-researched from an actual journalist?”

Pangram’s AI detection system is essentially a large machine learning model that was trained on tens of millions of known human documents. The startup then created a “synthetic mirror” for each document, replicating the topic, length, and tone of voice, but written by a frontier LLM. 

“Our model is learning the stylistic differences and the choices that AI makes consistently and is able to use that to learn what makes something AI-generated with high confidence,” Spero said, adding that the AI detector isn’t relying on copy-paste metadata or hidden watermarks. 

For Pangram, AI detection isn’t just about whether or not a piece of text was written entirely by AI. It’s also about distinguishing between levels of AI assistance — like in the case of someone who writes something themselves, but then asks AI to edit or clean it up. Spero believes AI assistance can be acceptable, just so long as the writer discloses their use of AI. 

As AI content floods the internet, Pangram raises M to detect it | TechCrunch
New York-based AI detection startup Pangram is on a mission to combat the AI slop infestation spreading across the internet, and it just raised  million on a bet that demand for tools that distinguish human-generated content from AI-generated text will only grow. 

Pangram’s fundraise — led by Menlo Ventures, with participation from Haystack, ScOp, Script Capital, and Cadenza — comes as the startup also launches its next-generation AI text detection model, Pangram 4, and an AI image detection model, Pangram Image. 







Pangram says the new text detection model is over 99% accurate at finding AI-assisted writing and mixed human-AI content, plus it can more easily detect AI humanizer programs. The AI image detector is only available via research preview for now; Pangram plans to release it more widely in the coming weeks.

Stanford AI and machine learning grads Max Spero and Bradley Emi launched Pangram about two years ago, after the launch of ChatGPT opened the floodgates for an internet full of bots, AI-generated SEO slop content, and what Spero calls “LLM-powered Russian disinformation campaigns and UAE-influenced campaigns on Twitter.”

“I think it’s just incredibly valuable to know whether what you’re looking at is something that’s AI-generated or not,” Spero told TechCrunch. “Especially text that you’re reading, because it changes how people approach the text. Is this something that I’m going to have to look out for hallucinations and jump in skeptically, or is this something that I trust was well-researched from an actual journalist?”

Pangram’s AI detection system is essentially a large machine learning model that was trained on tens of millions of known human documents. The startup then created a “synthetic mirror” for each document, replicating the topic, length, and tone of voice, but written by a frontier LLM. 

“Our model is learning the stylistic differences and the choices that AI makes consistently and is able to use that to learn what makes something AI-generated with high confidence,” Spero said, adding that the AI detector isn’t relying on copy-paste metadata or hidden watermarks. 


For Pangram, AI detection isn’t just about whether or not a piece of text was written entirely by AI. It’s also about distinguishing between levels of AI assistance — like in the case of someone who writes something themselves, but then asks AI to edit or clean it up. Spero believes AI assistance can be acceptable, just so long as the writer discloses their use of AI. 

Image Credits:Pangram

Pangram’s emergence comes at a time when AI usage is becoming more commonplace. In some cases, like the Canadian politician who read an AI prompt aloud in a speech to lawmakers, the mistakes result in ridicule. In other cases, as with certain lawyers making their case using fake citations created by ChatGPT, the consequences could be sanctions and fines.  

That backlash isn’t just costing individuals embarrassment or sanctions — it’s starting to show up in institutional rules, too.







The open-access archive arXiv introduced a new enforcement policy this year, stating that submissions containing evidence that authors failed to review LLM output (like hallucinated references or meta comments such as, “Would you like me to make any changes?”) can trigger a one-year submission ban.

Pangram isn’t the only one betting that AI detection will become more sought after. Competitors like Winston AI, Originality.ai, Copyleaks, and GPTZero are are chasing the same demand, each building its own detector.

Pangram’s technology, while not perfect, could help fuel the resistance to accepting the AI-generated content flooding the internet, the courtroom, and academic papers. 

Users can access Pangram via a -per-month subscription on the web or download the Chrome extension, which automatically labels posts in real time on X, LinkedIn, Substack, Reddit, and Medium. It also provides a feed health score with a percentage breakdown of human versus AI content on your screen. 

Pangram also offers its technology via API. Notably, Substack recently integrated Pangram’s technology into its platform to show readers which of their favorite authors write their newsletters using AI. Other API customers include Quora, schools and universities, publishers and agents, and recruiters, among others, per Spero. 

Does Pangram work?

Pangram detected AI-generated content even when lightly edited by a human. Image Credits:Pangram/TechCrunch

Spero said roughly one in 10,000 human documents are incorrectly labeled as AI with Pangram’s model, so I decided to put it to the test. The text detection model was very impressive but not perfect. It easily flagged entirely AI-generated news articles written by both ChatGPT and Claude, and was rarely fooled by my attempts to edit the AI-generated text into sounding more human. At the same time, Pangram did flag sentences that I completely rewrote as being AI-written. Pangram also wasn’t at all fooled by my attempts to prompt ChatGPT and Claude into evading AI detectors when generating content.  

I also gave ChatGPT and Claude one of my own articles and asked them to polish it up. Pangram gave it a 13% AI assisted score, which was probably close to accurate, but the model was able to detect subtle word-choice changes in some sentences and ignored them in others. It also flagged some sentences as AI-assisted when they were human written. That was notable because when I gave Pangram that same article in its entirety, as I had written it, it got a 100% human score. 

Maybe the problem was that news articles can be a bit dry and could easily sound like AI. So I tried a different tactic. I tested Pangram on my own more voicey, personal Substack newsletter content, pasting the first half of the text into Pangram and then asking ChatGPT and Claude to copy my style and write the second half. For the most part, Pangram easily detected human-written text versus AI-written text.







My limited testing of Pangram’s new image detection model turned out to be equally impressive. 

Image Credits:Pangram/TechCrunch

Pangram’s AI image detection system promises to spot AI-generated images across AI models, unlike OpenAI’s or Google DeepMind’s watermark-based checks, which mostly detect their own output. It works on pixel-level distributions, learning subtle statistical differences between real photos and AI-generated images. Spero says the model can even detect an AI image that appears inside a real-world photo. 

In my testing, the model easily detected AI-generated imagery, whether it was photorealistic or cartoonish. I can also confirm the model could detect an AI image appearing in a real-world photo — the heat map Pangram provides clearly lighting up over the image — though in one instance it incorrectly labeled a photo of an AI-generated image as human content. 

Spero says he doesn’t want his technology to fuel a witch hunt against people using AI for writing, but that there needs to be some sort of mechanism to push back against the slop. 

“The future that I see is that AI content just continues to proliferate,” Spero said. “We’re getting new GPUs faster than new people are being born. If we do not actively discriminate in favor of human content, then we’re just gonna see more and more AI, and it’s just gonna drown out any human signal that we have.”
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.#content #floods #internet #Pangram #raises #detect #TechCrunchai slop
Image Credits:Pangram

Pangram’s emergence comes at a time when AI usage is becoming more commonplace. In some cases, like the Canadian politician who read an AI prompt aloud in a speech to lawmakers, the mistakes result in ridicule. In other cases, as with certain lawyers making their case using fake citations created by ChatGPT, the consequences could be sanctions and fines.  

That backlash isn’t just costing individuals embarrassment or sanctions — it’s starting to show up in institutional rules, too.

The open-access archive arXiv introduced a new enforcement policy this year, stating that submissions containing evidence that authors failed to review LLM output (like hallucinated references or meta comments such as, “Would you like me to make any changes?”) can trigger a one-year submission ban.

Pangram isn’t the only one betting that AI detection will become more sought after. Competitors like Winston AI, Originality.ai, Copyleaks, and GPTZero are are chasing the same demand, each building its own detector.

Pangram’s technology, while not perfect, could help fuel the resistance to accepting the AI-generated content flooding the internet, the courtroom, and academic papers. 

Users can access Pangram via a $20-per-month subscription on the web or download the Chrome extension, which automatically labels posts in real time on X, LinkedIn, Substack, Reddit, and Medium. It also provides a feed health score with a percentage breakdown of human versus AI content on your screen. 

Pangram also offers its technology via API. Notably, Substack recently integrated Pangram’s technology into its platform to show readers which of their favorite authors write their newsletters using AI. Other API customers include Quora, schools and universities, publishers and agents, and recruiters, among others, per Spero. 

Does Pangram work?

Pangram detected AI-generated content even when lightly edited by a human. Image Credits:Pangram/TechCrunch

Spero said roughly one in 10,000 human documents are incorrectly labeled as AI with Pangram’s model, so I decided to put it to the test. The text detection model was very impressive but not perfect. It easily flagged entirely AI-generated news articles written by both ChatGPT and Claude, and was rarely fooled by my attempts to edit the AI-generated text into sounding more human. At the same time, Pangram did flag sentences that I completely rewrote as being AI-written. Pangram also wasn’t at all fooled by my attempts to prompt ChatGPT and Claude into evading AI detectors when generating content.  

I also gave ChatGPT and Claude one of my own articles and asked them to polish it up. Pangram gave it a 13% AI assisted score, which was probably close to accurate, but the model was able to detect subtle word-choice changes in some sentences and ignored them in others. It also flagged some sentences as AI-assisted when they were human written. That was notable because when I gave Pangram that same article in its entirety, as I had written it, it got a 100% human score. 

Maybe the problem was that news articles can be a bit dry and could easily sound like AI. So I tried a different tactic. I tested Pangram on my own more voicey, personal Substack newsletter content, pasting the first half of the text into Pangram and then asking ChatGPT and Claude to copy my style and write the second half. For the most part, Pangram easily detected human-written text versus AI-written text.

My limited testing of Pangram’s new image detection model turned out to be equally impressive. 

Image Credits:Pangram/TechCrunch

Pangram’s AI image detection system promises to spot AI-generated images across AI models, unlike OpenAI’s or Google DeepMind’s watermark-based checks, which mostly detect their own output. It works on pixel-level distributions, learning subtle statistical differences between real photos and AI-generated images. Spero says the model can even detect an AI image that appears inside a real-world photo. 

In my testing, the model easily detected AI-generated imagery, whether it was photorealistic or cartoonish. I can also confirm the model could detect an AI image appearing in a real-world photo — the heat map Pangram provides clearly lighting up over the image — though in one instance it incorrectly labeled a photo of an AI-generated image as human content. 

Spero says he doesn’t want his technology to fuel a witch hunt against people using AI for writing, but that there needs to be some sort of mechanism to push back against the slop. 

“The future that I see is that AI content just continues to proliferate,” Spero said. “We’re getting new GPUs faster than new people are being born. If we do not actively discriminate in favor of human content, then we’re just gonna see more and more AI, and it’s just gonna drown out any human signal that we have.”

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

#content #floods #internet #Pangram #raises #detect #TechCrunchai slop

Post Comment