Comparing Open AI and Google Deepmind: Who is Leading the AI Revolution?
Open AI’s ChatGPT has created a stir in the world of artificial intelligence. Now the questions turns to Google’s DeepMind — when will the tech behemoth put it to broad use? With the capacity for self-learning, DeepMind can stand on its own two feet. With the capability to revolutionize entire industries, it is essential for anyone passionate about AI to grasp DeepMind’s strengths. This article will delve into how DeepMind works, its goals and how it makes money.
The history of DeepMind
DeepMind Technologies was a British artificial intelligence laboratory launched in 2010. To generate finances, the business depended on investments from venture capitalists and entrepreneurs, such as Founders Fund and Horizons Ventures. This allowed them to expand their AI technology and hire new workers, reaching 75 employees prior to the Google acquisition of the company in 2014. The trio of founders Demis Hassabis, Shane Legg, and Mustafa Suleyman, alongside investor Jaan Tallin, began a venture with the aim of creating an “AI that thinks” — a pioneering effort in artificial intelligence.
Larry Wasserman revealed that the business had also obtained funds from several prominent billionaires. When Google purchased it, DeepMind was worth more than $500 million. Google’s ownership of DeepMind gave them entry to significant computing force, engineering proficiency, and consistent financial resources. Google has additionally consented to set up an ethical board to make sure DeepMind’s AI technology isn’t mishandled. After the purchase, DeepMind concentrated on deep learning and applied its AI innovation to other Alphabet ventures. As of December 2020, DeepMind depends upon a consistent stream of revenue from its parent firm Alphabet.
Technology
Google DeepMind has specialized in applying machine learning to develop functional algorithms for many uses. It focuses on using deep learning on a convolutional neural network, with a model-free reinforcement learning method called Q-learning. To achieve this, DeepMind inputs raw pixel data, allowing it to gain experience. It has also created a neural network that can play video games like a human. Furthermore, DeepMind has made a Neural Turing Machine, which consists of a neural network that can access an external memory similar to a conventional Turing machine and imitate the short-term memory of humans. DeepMind’s AI has been put into action on various video games from the 1970s and 1980s.
The organization has also constructed an app named Streams, which sends warnings to medical professionals about individuals in danger of suffering from acute kidney damage. This effort is part of their investigation partnership with the Cancer Research UK Centre at Imperial College London, aiming to strengthen the accuracy of breast cancer detection by utilizing machine learning on mammograms.
Additionally, they are working with the U.S. Department of Veterans Affairs to use machine learning to predict the onset of acute kidney injury in patients. Google has stated that DeepMind algorithms have greatly increased the efficiency of cooling its data centers. It has also collaborated with the Android team at Google for the creation of two new features which were made available to people with devices running Android Pie. The text-to-speech product, Cloud Text-to-Speech, released in 2018 is also based on WaveNet.
Difference between OpenAI and Google Deepmind
The differences between the technology applied by OpenAI and Google Deepmind are as follows:
• DeepMind has focused on establishing a baseline for Deep Reinforcement Learning algorithms, while OpenAI has focused on establishing a baseline for Deep Reinforcement Learning environments.
• DeepMind uses raw pixel data as input and learns from experience, using a deep learning on a convolutional neural network, with a model-free reinforcement learning technique called Q-learning.
• OpenAI’s language model GPT-3 uses AI to predict and complete lines of computer code, bringing the possibility of an AI system that can write itself one step closer.
• AlphaGo, developed by DeepMind, is the first computer program to defeat a professional human Go player and was deemed a decade ahead of its time.
• Google Brain and Microsoft could potentially be included in the top ranks of AI research output and impact, but Amazon and IBM are not quite on the same level.
And the winner is…
It’s hard to say which kind of AI technology will be most useful in the future. OpenAI and DeepMind have both accomplished great feats, and they will continue to innovate. AI has its advantages, but it could bring many dangers if not regulated properly. That is why safety and alignment in the development of AI is so important. If managed carefully and allowed to develop, both OpenAI and DeepMind may be essential forces in the progress of AI technology
