Applications
Infosys introduces open-source AI toolkit to boost trust and transparency
MUMBAI: Infosys is taking a crucial step toward making artificial intelligence safer, fairer, and more transparent. The global leader in digital services and consulting has officially launched the open-source Responsible AI Toolkit, a core component of the Infosys Topaz Responsible AI Suite. Designed to help enterprises innovate responsibly, the toolkit tackles AI ethics, security risks, and biases head-on.
The Infosys Responsible AI Toolkit builds on the company’s AI3S framework (scan, shield, and steer), providing enterprises with cutting-edge safeguards. It comes packed with specialised AI models and shielding algorithms that detect and mitigate threats such as privacy breaches, security vulnerabilities, biased outputs, deepfakes, hallucinations, and copyright violations. It also enhances AI model transparency, offering deeper insights into how AI-generated decisions are made—all without compromising performance or user experience.
By making the toolkit open-source, Infosys aims to foster collaboration and accelerate AI safety innovations across industries. The toolkit is fully customisable, supports various AI models and agentic systems, and integrates seamlessly across cloud and on-premise environments.
Infosys EVP, global services head, AI and industry verticals, Balakrishna D. R. (Bali) underscored the significance of this initiative, “As AI becomes central to driving enterprise growth, its ethical adoption is no longer optional. The Infosys Responsible AI Toolkit ensures that businesses remain resilient and trustworthy while navigating the AI revolution. By making the toolkit open source, we are fostering a collaborative ecosystem that addresses the complex challenges of AI bias, opacity, and security. It’s a testament to our commitment to making AI safe, reliable, and ethical for all.”
British High Commission head of science, technology and innovation Joshua Bamford highlighted the global significance of this initiative, “Infosys’ commitment to becoming an AI-first business and establishing the Responsible AI Office reflects bold innovation and ethical leadership. By going open source, Infosys is empowering enterprises, startups, and SMEs to leverage AI for groundbreaking advancements. Their Responsible AI Toolkit is a benchmark for technological excellence and when paired with a commitment to responsible practices and global sustainability can be an inspiring model for companies worldwide.”
Meanwhile, Meta public policy director – data economy and emerging tech, Sunil Abraham praised Infosys for driving open AI innovation, “We congratulate Infosys on launching an openly available Responsible AI Toolkit, which will contribute to advancing safe and responsible AI through open innovation. Open-source code and open datasets are essential to empower a broad spectrum of AI innovators, builders, and adopters with the information and tools needed to harness the advancements in ways that prioritise safety, diversity, economic opportunity, and benefits to all.”
The Indian government also lauded Infosys’ initiative, with government of India Ministry of Electronics and Information Technology (MeitY) additional secretary Abhishek Singh stating, “I am very happy to learn that Infosys has decided to open source their Responsible AI Toolkit. This will go a long way in making tools available for enhancing Security, Privacy, Safety, Explainability and Fairness in AI-based solutions and also help in mitigating bias in AI algorithms and models. This is critical for developing safe, trusted and responsible AI solutions. I am sure, startups and AI developers will greatly benefit from this Responsible AI Toolkit.”
With the Infosys Responsible AI Toolkit, enterprises can take a proactive approach to AI governance, ensuring that AI-driven systems remain ethical, fair, and secure. By going open-source, Infosys has set a new benchmark for corporate AI responsibility, encouraging organisations worldwide to collaborate, innovate, and drive AI adoption with trust.
Applications
Canva acquires animation and AI startups Cavalry and MangoAI
The deals strengthen Canva’s push into enterprise and AI-led design workflows
AUSTRALIA: Global visual communication platform Canva has stepped up its acquisition drive, buying UK-based 2D animation platform Cavalry and US-based AI startup MangoAI to deepen its AI-powered creative stack.
Cavalry, whose tools are used by brands including Amazon, Meta, Google and Netflix, will strengthen Canva’s motion design capabilities. The deal builds on Canva’s 2024 acquisition of Affinity, which has crossed four million downloads since launch. With Cavalry, Canva now counts seven Europe-based acquisitions, underscoring its global expansion strategy.
MangoAI, an early-stage startup focused on video advertising optimisation, will integrate its reinforcement learning systems into Canva AI. The move aims to enable brands to generate personalised marketing content in real time, cutting production cycles while improving campaign performance. MangoAI co-founder Vinith Misra will join Canva as reinforcement learning lead in its research lab.
Canva co-founder and chief operating officer Cliff Obrecht said the acquisitions reflect the company’s ambition to make professional-grade creative tools more accessible without sidelining human creativity. The goal, he said, is to bring everything from vector to motion design into a single, integrated suite.
The company now reports 265 million active users, including 31 million paid subscribers, and $4 billion in annualised revenue, up 36 per cent year on year. The latest buys further position Canva against rivals such as Adobe and Apple’s Creator Studio as it pushes deeper into enterprise workflows.
Canva head of pro design marketing Liam Fisher, said AI is intended to act as a creative assistant rather than a replacement, reinforcing the primacy of craft and individual design judgement.
