Hello reader,

Data masking used to be enough, but it's no longer effective.

Despite significant enthusiasm for AI, only 11% of enterprise CIOs have fully rolled out their AI initiatives. The main barriers? Data privacy and security concerns. Traditional data masking methods contribute heavily to these issues.

Masking reduces data quality. When data is masked, it becomes less accurate and less valuable for training AI models. Poor data quality directly leads to weaker AI outcomes, limiting effectiveness and innovation. According to Gartner, 85% of AI projects fail primarily because of fragmented or poor-quality data.

Additionally, traditional masking fails to adequately protect data. Even after masking sensitive information, advanced AI techniques can reconstruct or infer protected details, leaving data vulnerable to breaches. Gartner predicts that by 2027, over 40% of AI-related data breaches will arise from mishandling sensitive data within generative AI workflows.

The emergence of agentic AI workflows—where AI autonomously accesses sensitive data to execute complex tasks—exposes the weaknesses of traditional masking methods. Consider scenarios like an AI sales agent generating personalized customer interactions or an operational AI dynamically adjusting inventory. Traditional masking can't support these applications securely.

Confidential AI is designed to address these exact challenges. It safeguards data integrity and ensures strong privacy, enabling enterprises to securely leverage their most sensitive and valuable datasets—such as customer records, financial data, and proprietary insights—without risking compromise.

Major industry leaders—including Apple, Microsoft, Google, Amazon, and Meta—are adopting confidential AI solutions on their own clouds, clearly signaling a significant shift toward better privacy and superior data quality. If these companies recognize confidential AI as essential, shouldn't your organization?

Businesses adopting confidential AI experience fewer compliance challenges, reduce security risks, and speed up AI adoption. Confidential AI provides the clarity and security businesses need to maximize their AI investments effectively.

AI adoption is accelerating, and confidential AI is becoming essential—not optional—for safe, successful implementation.

Let's discuss how confidential AI can help your organization.

Aaron Fulkerson
CEO of OPAQUE

In the Lab

The Latest Happenings at Opaque Systems

Insights from Aaron Fulkerson at All Things Open AI and AI: Leaders Playbook
The market is moving fast—and confidential AI is the missing piece to unlock real innovation. Last week in North Carolina, OPAQUE was in the spotlight at key events. Aaron Fulkerson shared insights at All Things Open AI, the premier conference for AI technologists and innovators, presenting on deploying AI in the enterprise. At AI: Leaders Playbook, an executive event, Aaron was the featured speaker, emphasizing how organizations can protect sensitive data in AI workloads, without compromising performance or innovation. 

Thanks to Mark Hinkle, ATO organizers, and Triangle Technology Innovators Jason Johns for putting together such impactful events. Also, thank you, Nathaniel McCallum and the team from Microsoft Azure.

Whitepaper: Why Opaque? Why Now? 
AI-powered innovation is reshaping industries worldwide. Until now, breakthroughs have relied on training models on readily available public datasets. But the most valuable AI models of the future will be trained on sensitive, proprietary data to unlock predictive insights, drive automation, and improve decision-making. Check out our new whitepaper—Why Opaque? Why Now?—to uncover exactly how OPAQUE enables enterprises to fully leverage AI with sensitive data, and why it matters. 

Whitepaper: Bridging the Gap: How OPAQUE Complements Hyperscalers to Unlock Confidential AI
Hyperscalers provide confidential computing infrastructure but fall short of delivering the end-to-end privacy and compliance needs of enterprises. Read our new whitepaper—Bridging the Gap: How OPAQUE Complements Hyperscalers to Unlock Confidential AI—to learn how OPAQUE’s confidential AI platform bridges this gap, augmenting hyperscalers’ capabilities with a turnkey solution purpose-built for secure data and AI workloads.

Confidential Computing Summit Speakers Preview
We’re excited to welcome James Kaplan (Partner, McKinsey and CTO, McKinsey Technology) as a keynote speaker at Confidential Computing Summit 2025! With deep expertise in cloud transformation, cybersecurity, and technology operating models, Kaplan has guided Fortune 500 companies and private equity firms in maximizing their tech investments. And returning this year is Jason Clinton (CISO at Anthropic), who will share the latest insights on securing AI models and responsible AI deployment. Stay tuned for details on their 2025 sessions, along with the full speaker lineup. In the meantime: view the 2024 keynote presentations, register at the Early Bird rate, and book your hotel now to save—our hotel block is live, but rooms are limited!

Code for Thought

Worthwhile Reads

📝 Organizations are rewiring to capture AI’s value. Businesses are moving fast to unlock value from GenAI, with large enterprises leading the charge. According to the latest McKinsey Global Survey on AI, companies are redesigning workflows, appointing senior leaders to oversee AI governance, and addressing emerging risks as they scale AI across their business. More than 75% of respondents claim to use AI in at least one function, with GenAI adoption accelerating. As AI deployment grows, securing sensitive data to leverage AI’s full potential is critical, and OPAQUE streamlines this process.

🛡️ Mistakes companies make with cybersecurity. Cybersecurity is no longer just an IT concern—it’s a critical business priority. As cyber threats grow more advanced, businesses face rising risks from AI-powered attacks, insider threats, and unprepared workforces. Many still lack incident response plans and a company-wide culture of cyber readiness. Confidential AI solutions help organizations stay ahead, securing sensitive data and simplifying AI adoption so they can protect against evolving threats without compromise.

💼 Advancing agentic AI for the enterprise. At NVIDIA GTC, ServiceNow announced an expanded partnership with NVIDIA to advance agentic AI for the enterprise. The collaboration includes new evaluation tools that help businesses forecast and optimize agent performance before deployment. ServiceNow is also integrating NVIDIA’s Llama Nemotron models into its platform, enabling AI agents to better understand complex workflows and dynamically optimize business processes. The result: organizations can scale AI agents with greater confidence, driving more reliable automation and smarter outcomes from the start.

☁️ How AWS opened the floodgates for modern day cloud computing. As AWS grew into a household name, it paved the way for several industries and standards. Aaron Fulkerson shared that AWS helped propel the mobile application revolution that occurred shortly after it debuted its services. “Particularly in the early days of mobile application development, I don’t think anybody built on anything other than AWS,” Fulkerson said. “They all had their compute and storage on AWS.”