Google unveils Safety Charter for India's AI-led transformation

Tech giant Google unveiled its Safety Charter for India’s AI-led transformation, at the "Safer with Google India Summit".

The strategic blueprint operates through three foundational pillars —

i) keeping end users safe from online frauds and scams,

ii) strengthening cybersecurity for government and enterprise infrastructure, and

iii) building AI responsibly

Under its Digikavach program, Google has reached over 177 million Indians with AI-powered protections and awareness initiatives to combat financial fraud. AI integration across its platforms is transforming threat detection—Search now identifies 20x more scam-related pages; impersonation attacks on customer service and government sites have dropped by over 80 percent and 70 percent, respectively. Google Messages blocks over 500 million scam texts monthly and has issued more than 2.5 billion suspicious link warnings via on-device AI.

Since its October 2024 pilot in India, Google Play Protect has blocked nearly 6 crore high-risk app installation attempts across 13 million devices. Google Pay has issued 4.1 crore scam transaction alerts. Gmail continues to protect over 2.5 billion inboxes globally, automatically blocking more than 99.9 percent of spam, phishing, and malware.

Google is enhancing cybersecurity through an AI-first, secure-by-design approach focused on early threat detection and intelligence sharing.

Google.org has committed an additional $5 million to The Asia Foundation, expanding the APAC Cybersecurity Fund to support 10+ new cyberclinics, including partnerships with Indian universities to train MSMEs and students. In parallel, Google announced a collaboration with IIT-Madras in advancing Post-Quantum Cryptography, developing next-gen anonymous tokens that enable secure, privacy-first digital interactions for the future.

Preeti Lobana, vice president and country manager, Google India, said, "For India to become Viksit Bharat, we must build and maintain trust in the internet and our digital infrastructure. Trust is the bedrock of our digital aspirations and the reason India's digital economy has become an engine of growth. At Google, safety isn't an afterthought—it's embedded in our design principles, engineering processes, and company culture. Our AI systems constantly evolve to detect new threats and scams, even recognising malicious patterns in attacks that have never been seen before. This scalable capability helps us narrow or even eliminate the gap between defenders and attackers, which represents a huge leap forward in security."

Heather Adkins, vice president of Engineering, Google Security, highlighted, "India's digital journey continues to unlock incredible opportunities, but we also see the rise of sophisticated online threats evolving at machine speed. Our Safety Charter represents a comprehensive blueprint where AI isn't just narrowing the gap between attackers and defenders—it's eliminating it in some cases. AI has four incredible superpowers that are reversing the defender's dilemma with its reasoning, learning, speed and scale. We've used AI to discover previously unknown vulnerabilities in real-world software before attackers can exploit them. This defensive potential is game-changing, demonstrating how AI can serve as a tireless, transformative force keeping platforms and users secure."

Google's internal AI risk taxonomy promotes consistent risk understanding across teams, while AI-assisted red teaming uses adversarial testing to proactively detect and address harms.

SynthID has watermarked over 10 billion AI-generated assets with invisible digital signatures. In India, Google is fine-tuning inclusive AI via Gemini Language Testing and IndicGenBench, optimising performance across 29 Indic languages.

Read More: Google's grip loosens: Is AI the search disruptor?