Building trust in AI: The role of ecosystem partnerships and cybersecurity

Trust is the foundation of AI adoption 

At Google Cloud Next 2025, trust was a big topic. Trust needs to be the cornerstone of enterprise transformation. As companies accelerate AI journeys, the question is no longer “can we build it?” but “can we trust it?”. 

In this new era of intelligent systems, trust is built non the strength of the ecosystem and the security of the data that fuels it. 

Ecosystem partnerships: trust through collaboration 

AI innovation doesn’t happen in silos; it thrives in ecosystems. Google Cloud’s partnerships with industry leaders like Nvidia and Salesforce are not just about technology alignment. They’re about building a trusted, scalable foundation for enterprise AI.

“Google has been partnering with the biggest names in the industry  to create really a comprehensive environment and solution for the customer,” says Antoine Imbert, Group Head of Hyperscaler Partnerships at Capgemini. 

These partnerships are critical because trust is now the defining factor in AI adoption. According to McKinsey, trust is the foundation for the successful deployment of AI-powered products and services. If customers or employees don’t trust the outputs of AI systems, they won’t use them. This is especially true in enterprise environments, where explainability, transparency, and accountability are essential. 

By tapping into Google Cloud’s infrastructure and AI, Nvidia’s compute power, Salesforce’s customer intelligence, and Capgemini’s delivery expertise, enterprises are exploring: 

• Seamless, end-to-end AI journeys, from infrastructure to insight
• Accelerated innovations through co-developed, enterprise-ready solutions.
• Shared accountability, with multiple trusted leaders standing behind every deployment. 

Together, these partnerships form a powerful trust engine that enables businesses to innovate boldly, scale responsibly, and deliver AI that’s intelligent and dependable. 

But trust in AI doesn’t stop at collaboration. It must also be embedded in the systems that power and protect it. 

Cybersecurity: The invisible pillar of trust 

In the age of intelligent systems, trust begins with security. As AI agents become more autonomous and capable of interpreting data, making decisions, and executing tasks, the risks grow more complex. Enterprises need more than performance guarantees; they need assurance that their data, systems, and users are protected. 

Google Cloud has long been recognised for its leadership in cybersecurity. At Google Cloud Next 2025, this strength was very clear. From keynote sessions to the dedicated Security Hub,  the emphasis was on AI innovation being secure by design. 

“The importance and focus on cybersecurity, which has always been a great strength of Google, allows customers to really engage into AI with trust and with confidence that their data is safe,” says Antoine Imbert. 

It is more than just infrastructure. It is a new security paradigm. As outlined in Google’s latest Cloud CISO Perspectives report, securing AI agents requires layered, hybrid defences that combine traditional deterministic controls with dynamic, reasoning-based safeguards. 

Capgemini’s latest research reinforces this urgency. According to the Capgemini Research Institute, 92 percent of organizations experienced a cybersecurity breach in 2023, up from 51 percent in 2021. The rise of Gen AI has expanded the attack surface, introducing new risks such as prompt injection, deepfakes, and shadow AI. Yet, Capgemini also sees Gen AI as a powerful ally in defense. 

“Three in five organizations believe AI is essential to effective threat response, and 60 percent say they rely on AI to strengthen data, application, and cloud security.”  Generative AI in cybersecurity – Capgemini Research Institute 

Capgemini advocates for a comprehensive approach to AI security that includes: 

• Lifecycle protection: Securing every phase of Gen AI – from data collection and model training to deployment and monitoring.
• Governance frameworks: Establishing clear policies for ethical AI use, data handling, and employee training.
• AI-powered SOCs: Integrating AI into Security Operations Centers to reduce analyst fatigue, accelerate detection, and automate response.
• Proactive defense: Using Gen AI to simulate attacks, identify vulnerabilities, and train analysts with realistic threat scenarios.

Together, Google Cloud and Capgemini are helping enterprises move beyond reactive security to proactive resilience, embedding trust at the core of every AI initiative.