Helping your GenAI platform create content people want to actually watch, and taking down content that shouldn’t be anywhere.
San Francisco, CA
Joined July 2022
- Tweets160
- Following11
- Followers85
- Likes120
Raye and Michael B. Jordan are not dating. But photorealistic AI images of their supposed romance gathered 800,000 views on X before the platform even added an AI label.
As Luke told Vanity Fair: "The goal is to stop it before the toothpaste is out of the tube." Most of the time, the damage is already done before anyone gets there.
Most platforms respond to takedown requests quickly. X debates them in email threads, even for AI-generated child sexual abuse material. That is the accountability gap.
Full piece: vanityfair.com/story/michael…
#LotiAI #Deepfakes #GenerativeAI #DigitalRights #LikenessProtection
What does it actually look like to delete something from the internet? And what happens when the internet starts generating you without your permission?
Loti AI CEO Luke Arrigoni joined Jacob Goldstein on @pushkinpods What's Your Problem? to answer both.
Tens of thousands of takedowns a day, fully automated. Platforms cannot do it themselves because people do not trust them with their biometrics and they are not incentivized to police themselves.
As Luke put it: "It's hard for anyone to believe that they will properly gatekeep themselves."
On creativity: "The really cool stuff lives in the tails, and that's where humans have to push it."
Full episode: pushkin.fm/podcasts/whats-yo…
What happens when AI gets your identity wrong, and that version of you begins circulating online?
As AI systems ingest information from across the web, today’s fabrication can become tomorrow’s source material.
Lucy Boyle explores what that means for digital identity in @TheCut , with insights from Loti CEO @LukeArrigoni.
More: thecut.com/article/the-nepo-…
#Misinformation #GenAI #Tech #Deepfakes
Last year, we identified roughly 290 million instances of identity misuse, including 7 million deepfakes and 400,000 impersonator accounts.
Our CEO @LukeArrigoni joined @AdaptiveSec to discuss how AI is changing the scale of impersonation and what it takes to find and remove identity misuse across the internet.
Full interview: youtu.be/0ag2IKE_U3s?si=xS-D…
#LotiAI #Deepfakes #AI #GenAI
Cate Blanchett just launched a free consent registry at the European Parliament allowing anyone to define how AI can or cannot use their name, image, voice, and likeness.
The principle behind it is straightforward: "Your identity is your IP in the age of AI, and every person deserves the right to decide how AI can or cannot use it."
SZA discovered 200+ of her songs were used to train AI without her knowledge. McConaughey trademarked his voice. 400+ artists signed an open letter to the White House. The momentum is real.
Consent as a registry entry is a start. Consent built into the generation layer is what gives it teeth.
Full piece: euronews.com/culture/2026/06…
#LotiAI #GenerativeAI #DigitalRights #LikenessProtection #AIPolicy #ConsentInfrastructure
The race to build the best AI video model is largely over. The race now is which platform filmmakers actually want to work in every day.
By mid-2026, the industry has moved toward end-to-end workflows where scripting, visuals, audio, and editing happen in one system. Studios are reporting 85% faster previsualization and six-figure savings per project.
The framing is consistent across every platform: this is about increasing production value, not replacing filmmakers.
More: lnkd.in/g3db2Kf8
#LotiAI #GenerativeAI #Hollywood #DigitalRights #LikenessProtection #AIPolicy
Grimes launched Elf.Tech in 2023, allowing anyone to use her vocal likeness in exchange for a 50% royalty split. Her reasoning: voice cloning is already happening regardless of consent, so she would rather set the terms herself.
At Fortune Brainstorm Tech, she and Warner Music Group CEO Robert Kyncl agreed on something worth paying attention to. An abundance of AI-generated content will eventually push audiences to actively seek out human-made work, and mastery of storytelling will matter more, not less, as AI expands.
The artists defining what AI use looks like on their own terms are the ones who will be positioned when that correction arrives.
Full piece: fortune.com/2026/06/09/grime…
#LotiAI #GenerativeAI #LikenessProtection #NIL #MusicIndustry
Copyright protects what you make. It has never protected who you are as an artist.
The CREATOR Act, introduced this past June with Adobe's support, would change that. A federal right protecting visual artists' signature styles from intentional, commercial AI impersonation.
An AI platform used Philadelphia illustrator Fabiola Lara's style to showcase their own technology without asking. She had no legal recourse. That is the gap this bill is trying to close.
Full piece: blog.adobe.com/en/publish/20…
#LotiAI #CREATORAct #GenerativeAI #DigitalRights #AIPolicy
Warner Music Group and Suno's settlement is the first real blueprint for what licensed AI music looks like in practice.
Artists get the right to opt in for use of their names, images, likenesses, voices, and compositions in new AI-generated music, with revenue flowing back to them directly.
As Warner CEO Robert Kyncl put it: "AI becomes pro-artist when it adheres to our principles: committing to licensed models, reflecting the value of music on and off platform, and providing artists and songwriters with an opt-in."
Full piece: malaymail.com/news/showbiz/2…
#LotiAI #GenerativeAI #MusicIndustry #LikenessProtection #NIL #AIPolicy
The songs were fake. The streams were fake. The $8 million in royalties were real.
Michael Smith pleaded guilty to generating hundreds of thousands of AI songs and using bots to stream them billions of times, draining royalties from real artists and rights holders.
Every fraudulent stream represents more than illicit income, it represents value diverted away from real artists.
Full piece: techspot.com/news/111787-man…
#LotiAI #GenerativeAI #MusicIndustry #DigitalRights #StreamingFraud
Watermarking a deepfake tells you what was made. It does not stop it from being made.
Can you generate a deepfake of someone without typing their name? Yes. Easily.
Luke Arrigoni calls it a latent space attack. Describe everything around a person until the model fills in the center. No name needed.
"If I told you a man stranded on an island with a volleyball that turns into his best friend, you would probably immediately know who I'm talking about. And so would most models."
Name blocklists and watermarks are the wrong defense. The question that matters is whether content should be made at all.
Full interview: teapawt.org/MTSlive/status/2070596…
@MTSlive
#LotiAI #GenerativeAI #LikenessProtection #AIPolicy #Deepfakes
SITUATION EXPLAINED: Can you generate a deepfake of someone without ever typing their name?
We asked @LukeArrigoni, founder and CEO of @lotiapp
"If you watermark a deepfake, that's great, but it doesn't stop the fact that the deepfake is there. The watermarks are also invisible, so the damage could still be done."
"There's a mechanism before it gets watermarked where there's an allowance that says: should it be made at all?"
"What if someone says something tricky, someone that's really famous with a makeup brand that comes from a famous family that did a reality TV show. You could describe it and not use her name, and you can nudge the model to literally recreate her perfectly. You could even say the birth order and stuff like that."
"We call that the latent space attack. Latent space in a model is essentially where you surround the area of a person and describe everything around it... the inside part is the latent space."
"If I told you a man that's stranded on an island with a volleyball that turns into his best friend, you would probably immediately know who I'm talking about. And so would most models. But our system can detect that as well and then stop it."
YouTube is rolling out its AI deepfake detection tool to all creators 18 and over. It includes facial matching, identity verification, and a removal request flow built into YouTube Studio.
This is a meaningful step forward, but creators will increasingly need protections that span platforms, media formats, and the entire content lifecycle.
Full piece:
mobilesyrup.com/2026/05/18/y…
#LotiAI #YouTube #Deepfakes #GenerativeAI #LikenessProtection
Sons of Legion built a passionate community and AI scammers treated it as an opportunity.
50 to 60 fake accounts appear every day. Fans have sent thousands in crypto to impersonators. One person left a 40-year marriage and gave a scammer $50,000.
Loti AI has removed hundreds of impersonation pages for the band.
As Luke told TIME: "We're playing whack-a-mole at an industrial level."
Platforms remain the gatekeepers of where these scams spread. The technology exists. Collaboration between platforms, tech companies, and policymakers is the missing piece.
Full piece: time.com/article/2026/06/17/…
#LotiAI #Deepfakes #LikenessProtection #MusicIndustry #AIPolicy
AI is changing filmmaking, but creators shouldn't have to choose between participating in AI and protecting themselves from it.
This resonated across conversations at this year’s @Tribeca Storytelling Summit, creators deserve a say in how they participate.
That's the future we're building toward with Loti Interchange.
More: prnewswire.com/news-releases…
Kelsey Plum just became the first professional female athlete to launch a verified AI digital twin. Built on authorized recordings, consent-first terms, and direct oversight from the athlete herself.
The technology to replicate a public figure's voice and likeness already exists. The question is who controls the terms when it happens.
Full piece: fastcompany.com/91505719/wnb…
#LotiAI #GenerativeAI #LikenessProtection #NIL #WomenInSports
Over 200 music creators and members of Congress gathered in DC last month for GRAMMYs on the Hill, and AI was the topic at the center of every conversation.
Three bipartisan bills on the table: the NO FAKES Act, the TRAIN Act, and the CLEAR Act. Together they address one consistent principle: creators should control their identity, know how their work is being used, and be compensated when it is.
As Sen. Chris Coons put it, "Too many artists have had their intellectual property, likenesses and livelihoods stolen by online pirates and AI artists."
Full piece: recordingacademy.com/advocac…
#LotiAI #NOFAKESAct #GenerativeAI #DigitalRights #MusicIndustry #AIPolicy
"I think probably in the next year or two, it'll be impossible to distinguish what is fake or not."
Loti AI CEO @LukeArrigoni spoke with @CNET about deepfakes, digital identity, and why protecting human creativity and identity will be one of the defining challenges of the AI era.
Full piece: cnet.com/tech/services-and-s…
#LotiAI #AIPolicy #Deepfakes
Last week the Loti team was at the @Inc42 AI Summit in India, and it was exactly the kind of room our work was built for.
Our CTO Dr. Hirak Chhatbar joined a panel on algorithms, attention, and distribution in the AI era. The creator economy in India is moving fast, and the demand for trust, protection, and licensing infrastructure is growing with it.
As AI gets better at replicating voices and faces at scale, the question of who owns what becomes easier to ignore. Closing that gap is the work we show up for every day.
More: events.inc42.com/ai-summit/
#LotiAI #CreatorEconomy #LikenessProtection #Inc42AISummit
