A security startup that employs a zero-trust type of model described by one of its founders as a "need-to-know basis" for protecting personally identifiable information emerged from stealth mode this week with a $7.5 million seed funding round.
Skyflow — co-founded last year by former Salesforce vice president of platform Anshu Sharma and Prakash Khot, the former CTO of AthenaHealth and former senior vice president of engineering at Salesforce — has launched a cloud-based service that uses a group of secure application programming interfaces (APIs) that allow businesses to access and manage sensitive customer data such as Social Security numbers, credit card information, and healthcare data.
"What we keep seeing is people failing to protect the data, and the root cause of all of this is the obsession with compliance" only, says Sharma, co-founder and CEO of Skyflow. Existing approaches such as pure tokenization, hardware security modules, and various data encryption schemes require a team of experts to manage, he says. "You need an Ocean's 11."
Sharma says his company's approach was informed by technologies such as Apple's ApplePay that processes payment without storing credit card data in the cloud, as well as Twilio's secure APIs for telephony, for example. Skyflow's platform, he says, provides an end-to-end service that includes logging, access control, and data residency or storage, and polymorphic encryption.
The service stores the data in secure database "vaults" and ensures that when the data is shared among applications, only the necessary information is shared and it's anonymized and masked.
"Technically, we are like a database-as-a-service," he explains, with options for businesses to maintain their own encryption keys to the database instance, or Skyflow can manage them. "The net net is that it's a dedicated environment that has data layers controlled by them and code and management done by us."
Skyflow's encrypted database can handle Social Security numbers and credit card numbers, for example, using tokens that are homomorphically encrypted so that "internal applications never need access to the real data that's encrypted," he explains. This allows a third-party payment process, for example, to check a bank balance without needing or seeing whose account it was, he says.
A new study released this week from identity security firm Okta found a gradually growing interest in API security worldwide for businesses digitizing their operations with supply chains and other third parties. Some 21% of organizations say they plan to secure their APIs, with 40% in Europe and 30% in Australia and New Zealand, the study found.
With data breaches showing no signs of abating, data protection technology has been evolving, including beyond pure encryption and tokenization to more flexible and easier-to-manage approaches. "When organizations seek to protect their internal, sensitive data assets, an often-overlooked factor is the material impact it will have on their internal workers. Yes, you can block access to everything and call it a day, making it 'secure.' But how will that affect your data analysts? Or self-service analytics users?" says Paige Bartley, senior analyst at 451 Research, now part of SPG Global Market Intelligence. "Technologies that facilitate the secure use of sensitive assets let business users go about their jobs with minimum frustration and friction."
Bartley says approaches such as homomorphic encryption, for example, give businesses enough information from the data, without exposing the sensitive details. "If you frustrate the average workers within your organization, they will devise maladaptive work-arounds that undermine the security effort as a whole," she notes.
Startup ShardSecure, which launched last year, employs a data-shredding approach to securing sensitive data, for example. It automatically "shreds," mixes, and moves data such that it's unreadable if it lands in the wrong hands. Bob Lam, CEO and co-founder of ShardSecure, says his approach is focused on data security at the data layer.
451's Bartley notes that decentralized approaches to managing sensitive data can help protect it from attackers because "it is much easier to breach a single monolithic database than it is to crack dozens or even hundreds of decentralized repositories."
The trade-off, however, is that you may not have the full view of the data you once had. "A 'single pane of glass' view of data is typically easier to monitor for risk and security threats," she notes.
"Anshu and Prakash recognized the power and potential of the cloud while we were all at Salesforce," Todd McKinnon, CEO of Okta, said in a statement. "Every business needs to be digital to compete, and trust has become so core to building a brand. Digital privacy along with identity and security will only become more important as organizations evolve and grow online."
Skyflow's seed round was led by Foundation Capital's Ashu Garg, and joined by former CEO of GE Jeff Immelt and former AthenaHealth CEO Jonathan Bush. The company has hired experts in encryption, databases, key management, cloud security, and tokenization from Microsoft, Salesforce, Oracle, PayPal, and Cohesity, it said.
- How Cybersecurity Incident Response Programs Work (and Why Some Don't)
- The Problem with Automating Data Privacy Technology
- How Data Breaches Affect the Enterprise
- Top 10 Cyber Incident Response Mistakes and How to Avoid Them