Corporate US businesses safeguard themselves from hackers and pretenders with the aid of Doppel AI models, offering protection.
Rewritten Article:
Fighting Fire with AI: Doppel's Revolutionary Approach to AI-Driven Fraud
In the cat-and-mouse game between cybercriminals and security firms, Kevin Tyan believes the key lies in using AI to combat AI-driven fraud. As the founder of Doppel, a startup dedicated to protecting against psychological attacks, Tyan employs similar AI systems to help businesses mount swift and effective counterattacks against ever-evolving cyber schemes.
Every day, Doppel's ensemble of AI agents scours the internet, dark web, and social media platforms, analyzing 100 million potential threats. From fake websites and user accounts to fraudulent ads on popular platforms like Google, Instagram, and YouTube, these AI agents flag anything suspicious, leaving only the harmless or dangerous elements for human review.
According to Tyan, the system's accuracy stands around 90%, a number he claims is continuously growing. "If bad guys can use AI for just a few cents per dollar earned to launch attacks, then we must also be prepared to invest similar resources," says Tyan, speaking with our website USA.
In May 2023, Doppel announced it had raised $35 million in a funding round led by Bessemer Venture Partners, bringing its total funding to $55.5 million, valuing the company at $205 million. Originally co-founded by Tyan and technical director Rahul Madduluri in 2022 during their stint on Uber's flying car project, Uber Elevate, the new funding marks a significant milestone for the company.
Initially, Doppel aimed to combat NFT fraud and help crypto firms track and detect counterfeits, but in 2023, it expanded its client base to various industries. To identify potential threats and separate the wheat from the chaff, Doppel initially relied on contractors in countries such as the Philippines and India. However, by September 2024, the company had discovered that new OpenAI models with "reasoning" functionality could effectively handle the same tasks, leading to the replacement of contractors with AI agents. These agents enabled Doppel to automate 30% of its security operations.
"A team of human employees simply cannot process millions of notifications each day, so now these decisions are made by the AI agent," says Tyan, explaining the shift.
Productivity platform Notion faced a massive wave of attacks in 2023, with cybercriminals distributing fake ads among corporate clients and impersonating its CEO on social media. To combat these attacks, Notion turned to Doppel for assistance. Using its AI agents, Doppel swiftly intervened, eliminating thousands of elements related to the campaigns, and the frequency of attacks on Notion has since decreased.
According to Daniel Puukonen, head of security at Notion, "We've become a very expensive target."
Doppel's secret weapon is its "threat graph," a unique mapping tool that connects relationships and interactions between psychological attack campaigns, including phone numbers, IP addresses, and advertising accounts. With this tool, Doppel can better track AI-using hackers and provide enterprises with protection against future attacks.
"We want to give the good guys the same overview, so we don't have to fight individual cases, but can shut down the infrastructure of the attackers altogether," emphasizes Tyan.
To date, Doppel has attracted approximately 180 corporate clients, including United Airlines, OpenAI, Coinbase, and entities from the oil and gas sector, finance and insurance industries. Clients also benefit from a complex of industry-specific AI agents provided by Doppel. In the words of Bessemer Venture Partners partner Elliot Robinson, "We were impressed by their reach of first-tier clients beyond the IT industry."
As cybercriminals have become adept at using AI for phishing emails with dangerous links, Doppel's system confronts an array of new tactics. These include messages from fake CEO accounts requesting urgent money transfers, fraudulent LinkedIn profiles collecting private data, and even convincing corporate site traps. In 2024, Google removed 40 million fraudulent advertising accounts from its platform, many of which utilized AI-generated images and audio recordings of public figures and businesspeople to deceive and promote fraudulent schemes. Microsoft, meanwhile, thwarted attempted frauds that could have resulted in $4 billion losses for its clients, blocking 1.6 million bot registration attempts on its Azure platform.
Looking ahead, Tyan likens the current dilemma to that faced by Tom Cruise in the film "Mission: Impossible - Rogue Nation." "In that film, the AI is the antagonist. It can impersonate anyone, generate fake news, manipulate entire societies through false narratives in the digital space," reflects the entrepreneur. He views Doppel's mission as an opportunity "to build a company that solves one of the major problems in today's world - psychological attacks via AI."
Related Material
Enrichment Data:
Overall:
Doppel's AI-based system combats AI-driven fraud by leveraging autonomous AI agents that scan the internet, dark web, and social media platforms to identify fraudulent activities such as fake accounts and malicious advertisements. Doppel's unique "threat graph" allows it to connect these threats and reveal the malicious infrastructure behind attacks. By integrating large language models (LLMs) and human analysis, Doppel's platform, Doppel Vision, provides comprehensive visibility and takedown capabilities across various attack surfaces. Doppel serves various industries, including financial institutions, multinational corporations, media organizations, government agencies, and social media companies.
- To combat AI-driven fraud, Doppel, a startup valuing $205 million after raising $35 million in 2023, employs AI agents that analyze potential threats on the internet, dark web, and social media platforms, with an accuracy of approximately 90%.
- In response to the increasing use of AI by cybercriminals for phishing emails, dangerous links, fake CEO accounts, and fraudulent LinkedIn profiles, Doppel's complex of AI agents, including those leveraging large language models, offers a comprehensive solution to protect enterprises across various industries.
- Doppel's innovation extends beyond fraud detection, as it also equips its clients with a unique "threat graph" that connects relationships and interactions between psychological attack campaigns, providing enterprises with valuable insights for future protection against AI-enabled threats.
