Communiqués de presse
IBM lance AI OpenScale pour stimuler l'adoption et la transparence de l'intelligence artificielle
IBM’s new technology platform is the first of its kind. It will enable companies to manage AI transparently throughout the full AI lifecycle, irrespective of where their AI applications were built or in which environment they currently run. AI OpenScale can detect and address bias across the spectrum of AI applications, as those applications are being run.
As part of AI OpenScale, IBM also will debut NeuNetS, a major scientific breakthrough in which AI builds AI –making it possible to create complex, deep-neural networks from scratch. This will help address the AI skills gap and accelerate scaling AI.
“To accelerate AI adoption, businesses need one place to run all their AI,”said David Kenny, senior vice president, IBM Cognitive Solutions. “We are committed to enabling our clients to use the AI of their choice, regardless of vendor –and use IBM AI OpenScale to make it transparent and easy to manage. Only through enabling businesses to trust and scale their AI will we enable the AI economy.”
A recent study of 5,000 C-level executives by IBM shows that 82 percent of businesses surveyed are considering using artificial intelligence, but 63 percent lack the in-house talent to confidently manage the technology. Also, 60 percent are hindered by concerns over trust and compliance.
“AI is a significant enabler of business transformation and leaders need to demonstrate they trust and responsibly use AI. This includes an explanation of how AI capabilities are free of bias, protected from misuse, and consistently and predictably perform. For KPMG, it’s critical we have AI technology that helps facilitate trust among organizations, and strategically supports enterprise-wide transformation. IBM's new AI OpenScale capabilities are designed to enable and enhance trust amongst organizations,”according to Vinodh Swaminathan, Principal in KPMG’s Innovation & Enterprise Solutions group.
The AI OpenScale Advantage
AI OpenScale automates explainability, mitigates bias and provides auditability throughout the lifecycle of AI in a vendor-agnostic way. It helps companies:
· Understand how AI applications reach decisions –AI OpenScale explains how AI recommendations are made in everyday business terms.
· Address bias in AI applications automatically –AI OpenScale continually monitors AI applications and prevents bias through a unique, powerful and automated de-biasing technology.
· Ensure AI applications are auditable –AI OpenScale logs every prediction, every model version, and all the training data used, together with all metrics to help businesses comply with regulations such as GDPR.
· Use AI to build AI –AI OpenScale addresses the short supply of human AI engineers by throwing AI at the problem. IBM’s Neural Network Synthesis Engine (NeuNetS) will allow businesses to rapidly and automatically build neural networks –essentially to run AI –from scratch. NeuNetS will initially be available in AI OpenScale in beta.
· Manage and scale AI in an interoperable fashion – AI OpenScale can work with open source machine learning or deep learning models, such as Tensorflow, Scikitlearn, Keras and SparkML. It can also handle applications and models trained and hosted on common environments, including IBM Watson, IBM PowerAI, Seldon, AWS SageMaker, AzureML and other non-IBM engines.
AI OpenScale will be available via the IBM Cloud and IBM Cloud Private later this year.
About IBM & Artificial Intelligence A world leader in AI software, services, and technology for business, IBM has deployed Watson solutions in thousands of engagements with clients across 20 industries and 80 countries. IBM's Watson solutions are widely used in industries, including by 7 of the 10 largest automotive companies and 8 of the 10 largest oil and gas companies. IBM also built the IBM POWER9 processor from the ground-up for enterprise AI, which is at the heart of the world's smartest supercomputers, Summit and Sierra. Additionally, IBM Research is a world leader in the science of AI. In 2017, IBM secured 1400 AI-related patents. And, IBM recently revealed its leading-edge Project Debater, created by IBM Research scientists.
Contact(s) relations externes