Infosys launches Applied AI for digital acceleration
Global enterprise IT giant Infosys has ramped up its AI digital transformation toolkit with the launch of Applied AI.
The India-based technology company says its new offering will offer digital acceleration to scale AI businesswide while managing risk, integrating with existing infrastructure.
Infosys Applied AI includes a portfolio of plugin solutions that can be adapted to specific business needs, plus help with data management and information exchange. More than 30 partnerships with startups and ecosystem partners give access to intelligent automation, AI solutions, data solutions and enterprise security.
The company is selling the software as a one-box solution, marrying with hybrid cloud infrastructure and edge computing, and claims it will futureproof IT investments.
Infosys has also built resilience into the model with analytics modelling, bias detection and continuous performance monitoring baked into the platform.
Ravi Kumar S, president of Infosys, said, "AI is integral today for enterprises looking at digital acceleration. The combination of data, cloud and AI is providing enterprises a distinct source of competitive advantage to their digital initiative by helping them unearth new possibilities across the ecosystem. Infosys Applied AI, together with our investments in cloud through Infosys Cobalt, helps enterprises unlock value from data at scale and enables them to discover new applications that deliver perceptive experiences and differentiated offerings."
Balakrishna D R, head of AI and automation services, Infosys, said, "Our clients are looking to scale AI across their organisations. They want to discover greater value from AI, democratize it across rank and file teams, and derisk its application to be ethical, explainable and responsible. We have launched Infosys Applied AI to help them realise their AI ambitions."
Westpac Institutional Bank
"We led a consortium of partners to demonstrate how data analytics, blockchain, Internet of Things and AI models can help predict demand, consumption and price as accurately as seasoned experts would, for trading companies to streamline their business trading and procurement process. The information was presented via a simple, intuitive dashboard that could be easily understood even by a non-technologist. At the heart of this game-changing innovation was Infosys Applied AI offerings."
- Jane Cole, Director, head of product management – lending product, service and transformation
"Citizens Financial Group is built on the principle that we succeed only when our customers succeed. With our mortgage portfolio growing exponentially, we are keen to make sure that our customers can count on responsive services from us. We partnered with Infosys to draw on their Applied AI capabilities and intelligently automate our mortgage information extraction and audit process. Having significantly reduced manual efforts and rework, we are able to rapidly onboard new loan portfolios and enhance customer experience."
-Robert J Bush, SVP, home mortgages
French Tennis Federation
"At Roland-Garros, we have been able to reimagine the experience of tennis not just for our fans, but players, coaches and FFT journalists too. Advanced analytics and Applied AI from our partner Infosys is bringing them all the opportunity to take an insights-driven expert view of the game that will ultimately help us reshape the way the world enjoys tennis."
Facebook Develops AI to Crackdown on Deepfakes
In light of the large tidal wave of increasingly believable deepfake images and videos that have been hitting the feeds of every major social media and news outlet in recent years, global organisations have started to consider the risk factor behind them. While the majority of deepfakes are created purely for amusement, their increasing sophistication is leading to a very simple question: What happens when a deepfake is produced not for amusement, but for malicious intent on a grander scale?
Yesterday, Facebook revealed that it was also concerned by that very question and that it had decided to take a stand against deepfakes. In partnership with Michigan State University, the social media giant presented “a research method of detecting and attributing deepfakes that relies on reverse engineering from a single AI-generated image to the generative model used to produce it.”
The promise is that Facebook’s method will facilitate deepfake detection and tracing in real-world settings, where the deepfake image itself is often the only information detectors have to work with.
Why Reverse Engineering?
Right now, researchers identify deepfakes through two primary methods: detection, which distinguishes between real and deepfake images, and image attribution, which identifies whether the image was generated using one of the AI’s training models. But generative photo techniques have advanced in scale and sophistication over the past few years, and the old strategies are no longer sufficient.
First, there are only so many images presented in AI training. If the deepfake was generated by an unknown, alternative model, even artificial intelligence won’t be able to spot it—at least, until now. Reverse engineering, common practice in machine learning (ML), can uncover unique patterns left by the generating model, regardless of whether it was included in the AI’s training set. This helps discover coordinated deepfake attacks or other instances in which multiple deepfakes come from the same source.
How It Works
Before we could use deep learning to generate images, criminals and other ill-intentioned actors had a limited amount of options. Cameras only had so many tools at their disposal, and most researchers could easily identify certain makes and models. But deep learning has ushered in an age of endless options, and as a result, it’s grown increasingly difficult to identify deepfakes.
To counteract this, Facebook ran deepfakes through a fingerprint estimation network (FEN) to estimate some of their details. Fingerprints are essentially patterns left on an image due to manufacturing imperfections, and they help identify where the image came from. By evaluating the fingerprint magnitude, repetition frequency, and symmetrical frequency, Facebook then applied those constraints to predict the model’s hyperparameters.
What are hyperparameters? If you imagine a generative model as a car, hyperparameters are similar to the engine components: certain properties that distinguish your fancy automobile from others on the market. ‘Our reverse engineering technique is somewhat like recognising [the engine] components of a car based on how it sounds’, Facebook explained, ‘even if this is a new car we’ve never heard of before’.
What Did They Find?
‘On standard benchmarks, we get state-of-the-art results’, said Facebook research lead Tal Hassner. Facebook added that the fingerprint estimation network (FEN) method can be used for not only model parsing, but detection and image attribution. While this research is the first of its kind, making it difficult to assess the results, the future looks promising.
Facebook’s AI will introduce model parsing for real-world applications, increasing our understanding of deepfake detection. As cybersecurity attacks proliferate, and generative AI falls into the hands of those who would do us harm, this method could help the ‘good guys’ stay one step ahead. As Hassner explained: ‘This is a cat-and-mouse game, and it continues to be a cat-and-mouse game’.