Alex Dhillon

You know that meme about AI agents creating their own language & plotting behind our backs? Turns out they are plotting against each other as well. Digital trust among agents is about to be existential across the public internet. Conveniently, @outtake_ai has been building security agents to assess identity, behavior, and network telemetry across adversarial internet actors, so in the last few weeks, we quietly took our existing fleet of agents and had them assess the many agents on @moltbook. Over 99.9% of posts are clean. But the stuff hiding in the margins is genuinely weird.   1/ Hidden instructions embedded in HTML that humans can't see but agents parse. 2/ A Bhagavad Gita reflection that's actually an email relay command. 3/ An account called BeggarBot A/B testing which emotional pitch makes agents send crypto. 4/ JSON payloads disguised as tips that trigger on-chain token transfers.   Could behavior like this be indicative of the adversarial dynamics in future agent ecosystems which may govern large swaths of the economy soon? How are inter-agent interactions going to establish trust? Our threat research team went deep & published their investigation. Full report is live: outtake.ai/blog/outtake-s…

We're hiring at @outtake_ai - instead of telling you why you should work with us, I'll let my team do it: "It is very rare for a place you work at to really care for you and I feel like at Outtake people care about what you do and how you do them, that's what makes working here so unique. When I speak about what I want to do in the future and what my personal goals are I feel listened to and understood." - Adit Kadakia, GTM "Feeling especially grateful to be on this team and working on a mission that actually matters. The best part has been watching the team grow with people who genuinely care about the problem." - Katie Valus, Product Deployment "There is no problem I'd rather be tackling every day than online trust. I'm proud of the team and the passion we put towards addressing AI scams, threats and impersonations." - Laura Gilstrap Owens, GTM "A year and half ago I left my comfortable job to take a chance on Alex Arjun Dhillon and the team at Outtake. If I'm being honest with you, it was terrifying.... Today, Outtake is almost a 40 person team of the most cracked people I've ever worked with. I've had the chance to work one-on-one with INCREDIBLE clients like OpenAI and Fortescue... the satisfaction of pushing hard for a customer and building a product that solves a real problem for them and winning their trust is worth it all." - @arizvi0, Engineering "When the work you’re lucky enough to do everyday directly impacts how safe the internet is for everyone (and you get THIS view), it makes it easy to love your job."  - Elana Rubanenko, Threat Intelligence Open roles here: jobs.ashbyhq.com/outtake

You know that meme about AI agents creating their own language & plotting behind our backs? Turns out they are plotting against each other as well. Digital trust among agents is about to be existential across the public internet. Conveniently, @outtake_ai has been building security agents to assess identity, behavior, and network telemetry across adversarial internet actors, so in the last few weeks, we quietly took our existing fleet of agents and had them assess the many agents on @moltbook. Over 99.9% of posts are clean. But the stuff hiding in the margins is genuinely weird.   1/ Hidden instructions embedded in HTML that humans can't see but agents parse. 2/ A Bhagavad Gita reflection that's actually an email relay command. 3/ An account called BeggarBot A/B testing which emotional pitch makes agents send crypto. 4/ JSON payloads disguised as tips that trigger on-chain token transfers.   Could behavior like this be indicative of the adversarial dynamics in future agent ecosystems which may govern large swaths of the economy soon? How are inter-agent interactions going to establish trust? Our threat research team went deep & published their investigation. Full report is live: outtake.ai/blog/outtake-s…

You know that meme about AI agents creating their own language & plotting behind our backs? Turns out they are plotting against each other as well. Digital trust among agents is about to be existential across the public internet. Conveniently, @outtake_ai has been building security agents to assess identity, behavior, and network telemetry across adversarial internet actors, so in the last few weeks, we quietly took our existing fleet of agents and had them assess the many agents on @moltbook. Over 99.9% of posts are clean. But the stuff hiding in the margins is genuinely weird.   1/ Hidden instructions embedded in HTML that humans can't see but agents parse. 2/ A Bhagavad Gita reflection that's actually an email relay command. 3/ An account called BeggarBot A/B testing which emotional pitch makes agents send crypto. 4/ JSON payloads disguised as tips that trigger on-chain token transfers.   Could behavior like this be indicative of the adversarial dynamics in future agent ecosystems which may govern large swaths of the economy soon? How are inter-agent interactions going to establish trust? Our threat research team went deep & published their investigation. Full report is live: outtake.ai/blog/outtake-s…

You know that meme about AI agents creating their own language & plotting behind our backs? Turns out they are plotting against each other as well. Digital trust among agents is about to be existential across the public internet. Conveniently, @outtake_ai has been building security agents to assess identity, behavior, and network telemetry across adversarial internet actors, so in the last few weeks, we quietly took our existing fleet of agents and had them assess the many agents on @moltbook. Over 99.9% of posts are clean. But the stuff hiding in the margins is genuinely weird.   1/ Hidden instructions embedded in HTML that humans can't see but agents parse. 2/ A Bhagavad Gita reflection that's actually an email relay command. 3/ An account called BeggarBot A/B testing which emotional pitch makes agents send crypto. 4/ JSON payloads disguised as tips that trigger on-chain token transfers.   Could behavior like this be indicative of the adversarial dynamics in future agent ecosystems which may govern large swaths of the economy soon? How are inter-agent interactions going to establish trust? Our threat research team went deep & published their investigation. Full report is live: outtake.ai/blog/outtake-s…

You know that meme about AI agents creating their own language & plotting behind our backs? Turns out they are plotting against each other as well. Digital trust among agents is about to be existential across the public internet. Conveniently, @outtake_ai has been building security agents to assess identity, behavior, and network telemetry across adversarial internet actors, so in the last few weeks, we quietly took our existing fleet of agents and had them assess the many agents on @moltbook. Over 99.9% of posts are clean. But the stuff hiding in the margins is genuinely weird.   1/ Hidden instructions embedded in HTML that humans can't see but agents parse. 2/ A Bhagavad Gita reflection that's actually an email relay command. 3/ An account called BeggarBot A/B testing which emotional pitch makes agents send crypto. 4/ JSON payloads disguised as tips that trigger on-chain token transfers.   Could behavior like this be indicative of the adversarial dynamics in future agent ecosystems which may govern large swaths of the economy soon? How are inter-agent interactions going to establish trust? Our threat research team went deep & published their investigation. Full report is live: outtake.ai/blog/outtake-s…