Meet the internet pioneer who declared war on spam and is helping AWS prepare for AI cyber threats
In 2026 Paul Vixie, an Internet Hall of Fame inductee, joined AWS as a Distinguished Engineer to build AI‑driven security. AWS now offers GuardDuty’s AI‑enhanced threat modeling and an automated remediation service that scans CloudFormation/Terraform for wildcard IAM permissions and can open pull requests in GitHub.
AWS Distinguished Engineer Paul Vixie started the first anti-spam company and helped scale the Domain Name System, which makes it possible to type web addresses as names instead of numbers.
Vixie created the first neutral commercial internet exchange in the U.S.—infrastructure that enabled early internet commerce and connectivity.
Vixie is now helping to advance security at AWS by identifying ways to protect against AI-enabled threats.You probably didn’t think much of it, but when you checked your email this morning, your inbox wasn’t full of spam. When you checked last night’s scores, you typed in the name of your favorite sports website, not an unwieldy string of numbers. Both email filters and domain names are essential conveniences of modern life. And both are in use at global scale thanks to the work of a guy named Paul Vixie.These days, Vixie is an Amazon Web Services (AWS) Distinguished Engineer, a select crew of executive-level mavericks with the autonomy to solve complex, ambiguous problems that have stumped everyone else. Or as Vixie puts it, experts whose job it is “to find trouble, and get into it.”From his days as a 16-year-old computer programmer to his Internet Hall of Fame career, Vixie has been motivated to find that kind of trouble. So, it’s no surprise that in 2026, he has turned his attention and abilities to the rise of agentic AI, thinking not just about how to secure AI but how to use the tech itself to make systems more secure.“Non-technical people are now talking about AI over dinner. Everyone has an opinion, everyone’s used an AI model on their phone or their browser, and so it won't just be enterprises and families who employ it to make their lives better,” Vixie says. “The ways in which AI can be used to cause harm are probably my biggest bugaboo right now. If I were a bad guy, what would I be leveraging this for?”
Inside Gen AI's split-second sprint from question to answer
AWS powers the work of “inference” with custom chips, smart routing systems, and purpose-built infrastructure—making AI faster and more affordable.But even though he’s thinking about all the nefarious ways AI can be used, he takes a broader perspective. “The challenge is not simply that it's a new thing we must figure out how to secure. We get the benefit of figuring out how we're going to use AI to secure other things, to identify and fix vulnerabilities.”As he sets out on the latest stage of his career, Vixie shared five lessons about life and leadership he’s learned along the way.
Analysis & Recommendations
Why This Matters
Sellers using AWS can leverage AI‑generated attack simulations—such as a stolen API key targeting an S3 bucket—to pre‑empt breaches. Automatic detection of over‑permissive IAM policies and auto‑generated pull‑request fixes reduce misconfigurations, lowering the risk of credential‑stuffing and data loss across storefronts.
Key Takeaways
2026: Paul Vixie became an AWS Distinguished Engineer focusing on AI cyber‑threats.
GuardDuty now includes an AI‑enhanced threat modeling feature that simulates attacks like stolen API keys on S3 buckets.
The AI remediation service automatically scans CloudFormation/Terraform templates for wildcard IAM permissions ("*") and can create GitHub pull req...
Remediation outcomes are fed back into the model via Security Hub, improving future alert relevance.
Recommended Actions
→Enable GuardDuty AI threat modeling: AWS Console > GuardDuty > Settings > toggle “AI‑enhanced threat modeling” on.
→Integrate the AI remediation service with your CI/CD: In AWS CodePipeline, add the “AI Remediation” action and authorize it to open pull requests i...
→Mark AI‑suggested fixes as resolved: Open Security Hub, locate the AI remediation finding, and select “Mark as Resolved” to feed the result back in...
Frustration is fuel (if you channel it)A day in the life of Paul Vixie begins as follows: “Every morning I wake up with a high likelihood of being angry about something. I guess the difference between me and most other people is that I go and do something about it.”Right now, he’s getting worked up about how bad actors might leverage AI. But in the beginning it was frustration at flunking high school (because he was too busy programming to do his homework) that led him to search classified ads for jobs that didn't require a college degree—or proof of age.Paul Vixie, AWS VP and Distinguished EngineerThat move launched a series of unorthodox career decisions. When, in the 1990s, he woke up angry about the amount of spam email he was receiving, he started the first anti-spam company. “We invented the first anti-spam technology, and it is still in universal use.”
Sometimes the best qualification is being the one who cares enoughVixie didn't invent the Domain Name System (DNS); he inherited the project from someone else. But he was the one person who came across the idea, saw what had to be done to make it commercially viable, and cared enough to do it. DNS is essentially the internet's address book, a system that translates web addresses like Amazon.com into sequences of numbers that computers understand. Without it, you'd need to memorize numerical strings like 205.251.242.103 just to check your email.
How one Amazon engineer is making the internet a safer place for all of us (and cats, too)
AWS VP and Distinguished Engineer Tom Scholl collaborates with networks across the globe to stop cyberattacks by tracking traffic from bad actors at its source.When, in his job as a programmer in the mid-1980s, Vixie encountered the main implementation of DNS (known as the Berkeley Internet Name Domain, or BIND), it had been abandoned. "It was just sitting there, so I grabbed the code and started fixing it." After mentioning what he was working on to someone, they asked for a copy. He obliged and began regularly publishing copies so anyone could use it. “I soon received a long list of feature requests, bug reports, and all kinds of other things,” he said.
Build for scale you can’t yet imagineWhen Vixie started his career, a computer that could be connected to the internet cost millions of dollars. There were so few internet users, you could keep everyone’s email address on one simple list. But he and others could see that connectivity would become a commodity, and that the list would get a billion times larger. So he went to work on ensuring Domain Name System tech could scale.The superpower of DNS was delegating naming authority so that, for example, Amazon could add new web destinations ending in Amazon.com without outside permission. “This is one of the main reasons DNS could scale, versus listing everything on the internet that you could want to connect to in a single file,” he said.And while Vixie understood the potential of DNS, he didn't necessarily foresee its importance in the era of agentic AI. Today, every time a model queries an external API, retrieves training data, or serves responses to users, DNS is behind the scenes, translating those requests into the numerical addresses that make communication possible.
Be a “relevance junkie”Being online in the early 1990s wasn't an immediate passport to speaking to everyone else on the internet. The reality was more limited. You had to pay telecommunications companies exorbitant fees to connect to people on separate networks. For a startup trying to build "Earth's biggest bookstore," those costs were potentially business-killing.As a 20-something programmer for a company that sold commercial data center equipment, Vixie was certain the internet was the future of technology, but many of his colleagues dismissed it as a playground for West Coast hippies. So he started a side project: connecting networks to each other in his employer's data center. No contracts, no permissions, just a gut feeling about what needed to exist.“I can get a pretty good feedback loop of building things that help other people to build things,” Vixie said. “I’m a relevance junkie.”Vixie at Amazon's Palo Alto officeHaving effectively created the United States’ first neutral and commercial internet exchange, which was critical in helping fledgling online companies like Yahoo, Amazon, and countless others get off the ground, he left his job. Three months later, his former boss called to say that they’d uncovered the covert enterprise and wanted him back as a consultant to turn it into a legitimate venture.“I knew the internet would lead to people building businesses in new ways,” said Vixie. “I didn't know what those businesses would be, but I believed they should have the freedom to do it.”
Stay curious about what needs doing todayAbout a year into his Distinguished Engineer role, AWS asked Vixie to lead the Office of the Chief Information Security Officer (OCISO) and serve as Deputy CISO. He agreed immediately. “It's very difficult to take a startup founder like me and get them to speak the words, 'Yeah, that's not what I came here for,' because whatever needs doing is what you came here for,” he said.“My career is from the ‘Journal of Irreproducible Results.’ I did what I did because I was standing nearby when things needed doing. The things other people do will be specific to the era they're working in, but the irreproducibility is, I think, essential.”
AWS CEO calls AI inference a new building block that transforms what developers can build
Task-accomplishing agents deliver more than just content generation, and enterprises will see massive returns in 2026, says AWS CEO Matt Garman.His advice: don't put yourself on rails toward a predetermined destination. “You’ve got to keep your options open and follow your passion. When you do that, you’ll create opportunities through the stream of your experiences, and they will end up being the right opportunities for you.”Vixie’s latest passion? Program correctness: using AI and modern computing to help ensure software behaves as intended. “It’s about catching potential issues early in the development process,” he explained.True to form, he's found new trouble to get into.Learn more about AWS Cloud Security and get the latest news in the AWS Security Blog.
Trending news and stories
Amazon introduces feeds to make it easier for merchants to reach more customers through AI-powered Shop Direct
Everything you need to know about Amazon Nova
Amazon introduces faster delivery with new 1-hour and 3-hour options
The top 7 questions about Kindle for kids, answered
Key takeaways
AWS Distinguished Engineer Paul Vixie started the first anti-spam company and helped scale the Domain Name System, which makes it possible to type web addresses as names instead of numbers.
Vixie created the first neutral commercial internet exchange in the U.S.—infrastructure that enabled early internet commerce and connectivity.
Vixie is now helping to advance security at AWS by identifying ways to protect against AI-enabled threats.You probably didn’t think much of it, but when you checked your email this morning, your inbox wasn’t full of spam. When you checked last night’s scores, you typed in the name of your favorite sports website, not an unwieldy string of numbers. Both email filters and domain names are essential conveniences of modern life. And both are in use at global scale thanks to the work of a guy named Paul Vixie.These days, Vixie is an Amazon Web Services (AWS) Distinguished Engineer, a select crew of executive-level mavericks with the autonomy to solve complex, ambiguous problems that have stumped everyone else. Or as Vixie puts it, experts whose job it is “to find trouble, and get into it.”From his days as a 16-year-old computer programmer to his Internet Hall of Fame career, Vixie has been motivated to find that kind of trouble. So, it’s no surprise that in 2026, he has turned his attention and abilities to the rise of agentic AI, thinking not just about how to secure AI but how to use the tech itself to make systems more secure.“Non-technical people are now talking about AI over dinner. Everyone has an opinion, everyone’s used an AI model on their phone or their browser, and so it won't just be enterprises and families who employ it to make their lives better,” Vixie says. “The ways in which AI can be used to cause harm are probably my biggest bugaboo right now. If I were a bad guy, what would I be leveraging this for?”
Inside Gen AI's split-second sprint from question to answer
AWS powers the work of “inference” with custom chips, smart routing systems, and purpose-built infrastructure—making AI faster and more affordable.But even though he’s thinking about all the nefarious ways AI can be used, he takes a broader perspective. “The challenge is not simply that it's a new thing we must figure out how to secure. We get the benefit of figuring out how we're going to use AI to secure other things, to identify and fix vulnerabilities.”As he sets out on the latest stage of his career, Vixie shared five lessons about life and leadership he’s learned along the way.
Frustration is fuel (if you channel it)A day in the life of Paul Vixie begins as follows: “Every morning I wake up with a high likelihood of being angry about something. I guess the difference between me and most other people is that I go and do something about it.”Right now, he’s getting worked up about how bad actors might leverage AI. But in the beginning it was frustration at flunking high school (because he was too busy programming to do his homework) that led him to search classified ads for jobs that didn't require a college degree—or proof of age.Paul Vixie, AWS VP and Distinguished EngineerThat move launched a series of unorthodox career decisions. When, in the 1990s, he woke up angry about the amount of spam email he was receiving, he started the first anti-spam company. “We invented the first anti-spam technology, and it is still in universal use.”
Sometimes the best qualification is being the one who cares enoughVixie didn't invent the Domain Name System (DNS); he inherited the project from someone else. But he was the one person who came across the idea, saw what had to be done to make it commercially viable, and cared enough to do it. DNS is essentially the internet's address book, a system that translates web addresses like Amazon.com into sequences of numbers that computers understand. Without it, you'd need to memorize numerical strings like 205.251.242.103 just to check your email.
How one Amazon engineer is making the internet a safer place for all of us (and cats, too)
AWS VP and Distinguished Engineer Tom Scholl collaborates with networks across the globe to stop cyberattacks by tracking traffic from bad actors at its source.When, in his job as a programmer in the mid-1980s, Vixie encountered the main implementation of DNS (known as the Berkeley Internet Name Domain, or BIND), it had been abandoned. "It was just sitting there, so I grabbed the code and started fixing it." After mentioning what he was working on to someone, they asked for a copy. He obliged and began regularly publishing copies so anyone could use it. “I soon received a long list of feature requests, bug reports, and all kinds of other things,” he said.
Build for scale you can’t yet imagineWhen Vixie started his career, a computer that could be connected to the internet cost millions of dollars. There were so few internet users, you could keep everyone’s email address on one simple list. But he and others could see that connectivity would become a commodity, and that the list would get a billion times larger. So he went to work on ensuring Domain Name System tech could scale.The superpower of DNS was delegating naming authority so that, for example, Amazon could add new web destinations ending in Amazon.com without outside permission. “This is one of the main reasons DNS could scale, versus listing everything on the internet that you could want to connect to in a single file,” he said.And while Vixie understood the potential of DNS, he didn't necessarily foresee its importance in the era of agentic AI. Today, every time a model queries an external API, retrieves training data, or serves responses to users, DNS is behind the scenes, translating those requests into the numerical addresses that make communication possible.
Be a “relevance junkie”Being online in the early 1990s wasn't an immediate passport to speaking to everyone else on the internet. The reality was more limited. You had to pay telecommunications companies exorbitant fees to connect to people on separate networks. For a startup trying to build "Earth's biggest bookstore," those costs were potentially business-killing.As a 20-something programmer for a company that sold commercial data center equipment, Vixie was certain the internet was the future of technology, but many of his colleagues dismissed it as a playground for West Coast hippies. So he started a side project: connecting networks to each other in his employer's data center. No contracts, no permissions, just a gut feeling about what needed to exist.“I can get a pretty good feedback loop of building things that help other people to build things,” Vixie said. “I’m a relevance junkie.”Vixie at Amazon's Palo Alto officeHaving effectively created the United States’ first neutral and commercial internet exchange, which was critical in helping fledgling online companies like Yahoo, Amazon, and countless others get off the ground, he left his job. Three months later, his former boss called to say that they’d uncovered the covert enterprise and wanted him back as a consultant to turn it into a legitimate venture.“I knew the internet would lead to people building businesses in new ways,” said Vixie. “I didn't know what those businesses would be, but I believed they should have the freedom to do it.”
Stay curious about what needs doing todayAbout a year into his Distinguished Engineer role, AWS asked Vixie to lead the Office of the Chief Information Security Officer (OCISO) and serve as Deputy CISO. He agreed immediately. “It's very difficult to take a startup founder like me and get them to speak the words, 'Yeah, that's not what I came here for,' because whatever needs doing is what you came here for,” he said.“My career is from the ‘Journal of Irreproducible Results.’ I did what I did because I was standing nearby when things needed doing. The things other people do will be specific to the era they're working in, but the irreproducibility is, I think, essential.”
AWS CEO calls AI inference a new building block that transforms what developers can build
Task-accomplishing agents deliver more than just content generation, and enterprises will see massive returns in 2026, says AWS CEO Matt Garman.His advice: don't put yourself on rails toward a predetermined destination. “You’ve got to keep your options open and follow your passion. When you do that, you’ll create opportunities through the stream of your experiences, and they will end up being the right opportunities for you.”Vixie’s latest passion? Program correctness: using AI and modern computing to help ensure software behaves as intended. “It’s about catching potential issues early in the development process,” he explained.True to form, he's found new trouble to get into.Learn more about AWS Cloud Security and get the latest news in the AWS Security Blog.
Trending news and stories
Amazon introduces feeds to make it easier for merchants to reach more customers through AI-powered Shop Direct
Everything you need to know about Amazon Nova
Amazon introduces faster delivery with new 1-hour and 3-hour options
The top 7 questions about Kindle for kids, answered
Comments
Join the discussion
Log in or create an account to share your thoughts on this update.
Comments
Join the discussion
Log in or create an account to share your thoughts on this update.
No comments yet. Be the first to share your thoughts!