“Artificial Intelligence will reach human levels by around 2029. Follow that out further to, sat, 2045, and we will have multiplied the intelligence – the human biological machine intelligence of our civilisation – A billion-fold.” – Ray Kurzweil, American inventor and futurist. Going by this quote, artificial intelligence has become less of a trend and more of a core ingredient in every aspect of computing. AI is already invading every industry vertical, right from helping doctors perform surgeries with robots to autonomous cars, and mimicking the way people speak. In just 2 years, artificial intelligence is likely to increase the rate of innovation in India by 230%. Though AI is just at the tip of the iceberg for many start-ups major tech giants from Google to Amazon and Microsoft to Facebook are already in a frenzied race to become leaders in artificial intelligence by dedicating resources to make remarkable breakthroughs in AI. As every AI company scrambles to infuse their products with human-like intelligence to provide next-level services and products, companies like OpenAI, Apple, Nvidia and DJI are AI trailblazers that are at the pinnacle of the AI revolution. Each AI company has their own secret sauce of success in the AI revolution. Here’s how each of these companies is embracing AI and betting big on artificial intelligence in this frenzied race-
The Next Great AI Company – Apple vs OpenAI vs NVIDIA vs DJI
Started In – 1977
Headquartered In – Cupertino, California
Employee Strength – 1, 23,000
Industry – Computer Hardware, Computer Software, Consumer Electronics, Financial Technology, Artificial Intelligence
AI Company – Artificial Intelligence at Apple: FaceID and Siri
With over 20 AI acquisitions since 2010, Apple is the first major tech company that has integrated a smart assistant into its OS. If you’ve got an iPhone, you already know about the “Apple Brain” inside it. When the Maps on your iPhone pop up to the restaurant location you’ve reserved for dinner, much before you even type it in. Or when the iPhone identifies a caller that is not in your contacts but you did send an email recently. These are all made possible by Apple’s adoption of AI, deep learning, and neural networks.
The 2017 A11 Bionic chipset that powers features like FaceID and the groundbreaking personal assistant Siri are Apple’s commitment to world-class AI. Apple’s FaceID technology lets users unlock their iPhone just by looking at it. This technology stems from Apple’s move in chips and computer vision that includes the acquisition of the AI company RealFace. The smart assistant on your iPhone will wake up when you utter the phrase “Hey Siri” but not when someone from your family or friends utters this phrase. This ensures that all iPhones in a place don’t respond when one person says the wake phrase “Hey Siri”. And yes it does not collate gigabytes of audio data to do this. Apple personalises Siri without hoovering upon data using a privacy-preserving machine learning method known as Federated Learning. A major factor that differentiates Apple from other tech companies in leveraging AI is its commitment to data privacy.
A12 Bionic Chipset- The Fastest CPU Ever in a Smartphone
With greater control and competence in optimising chips running the AI computations on its devices, iPhones today are powered with A13 Bionic chipset that helps understand the world around them using AI algorithms. The A13 chipset features powerful machine learning capability with an octa-core Neural Engine, works six times faster than the earlier chipsets, boosts performance by roughly 20% across the board and is capable of performing as many as one trillion operations per second. The iPhone camera uses machine learning to process images and the chip’s lightning-fast speed lets it shoot 4K videos at 60 fps with HDR enabled. It is the novel advancements in AI chips that allow the latest iPhones to provide augmented reality experiences and slicker camera effects. Apple aims to enable the new iPhones to process Siri requests offline with its recent acquisition in Xnor.ai which has an edge AI that runs with 15x times more memory and is 10x times faster than cloud-based systems.
Started In – 2015
Headquartered In – San Francisco, California
Employee Strength – 120
Industry –Artificial Intelligence
Recent Achievements – Built OpenAI Five, the first AI that defeats Dota 2( Dota is a game more challenging than chess) World champions in an esports game.
Working with a mission to make sure artificial general intelligence benefits all of humanity, OpenAI is trending the AI landscape for studying the ethics and safety of Artificial Intelligence. Here are some core developments at OpenAI that make it the most promising AI company –
- One of its recent AI developments is a new AI model Jukebox that generates genre-specific music. It generates music based on multiple factors like lyrics and instrumentation. OpenAI used CNN (convolutional neural networks) to train the model on a dataset of 1.2 million songs along with lyrics and metadata scraped from LyricWiki. The model was trained by compressing the audio clips and various snippets of lyrics and encoding them in a format that the neural network can understand. The model then generated new compressed audio using a transformer and is upsampled to generate raw audio.
- Dactyl, its humanoid robotic hand can solve a Rubik’s cube one-handed. Dactyl is a great leap forward in the AI space for the dexterity of robotic appendages. The humanoid robotic hand learns new tasks using virtual simulations before it is presented with an actual challenge to overcome. OpenAI aims to build a general-purpose robot that can actually perform a variety of real-world tasks without needing to be programmed particularly and without requiring training for months of real-world time.
- Another fine accomplishment by OpenAI is its language bot GPT-2 that is very good at generating paragraphs of convincing text based on the sample you throw at it , such as a paragraph of a fictional story, headline of a news article, or essay on a particular topic. GPT-2, the transformer-based language learning model was trained on a dataset of 8 billion webpages(approximately 40GB of data) from Reddit with 1.5 billion parameters making it flexible enough to work on diverse use cases.
Started In – 1993
Headquartered In – Santa Clara, California
Employee Strength – 13,227
Industry – Computer Hardware, Semiconductors, Consumer Electronics, Video Games
Recent Achievements – Mellanox acquisition closed on April 27, 2020 is NVIDIA’s move to establish an end-to-end AI stack and drive the growth of next-gen data centres by combining artificial intelligence with high-speed computing and data processing.
AI Company – Core AI Developments at NVIDIA
NVIDIA often thought of as a chip-making company for making video games cool, is now leading the way in AI computing with the use of GPU’s for areas like artificial intelligence and high-performance computing. NVIDIA’s core strength lies in its GPU which is the most important chip architecture for training and inferencing server and cloud-based applications of deep learning, machine learning, and NLP. Artificial intelligence depends on GPU acceleration and this AI company delivers it wherever you need it to be it laptops, desktops, cloud, supercomputers or the data centres.
- NVIDIA uses AI in a process known as Deep Learning Super Sampling to speed up the process of photorealism to render natural, realistic, and eye-pleasing experience. Unlike most companies that use raytracing ((rendering the model by bouncing light on it)for photorealism which is computationally heavy and expensive, NVIDIA employs AI to cut down the computation time and cost and improve performance. Nvidia’s RTX platform fuses ray tracing to help gaming companies bring photorealistic digital humans to life in real-time.
- NVIDIA also manufactures AI processors for optimised search engine recommendations, improved language comprehension, cloud, and other industries like warehousing, retail, and logistics. Nvidia’s Tegra processor segment is the money maker that focuses on products that integrate a full computer onto a single chip. This kind of AI processor finds extensive use in autonomous robots, vehicles, drones.
Started In – 2006
Headquartered In – Shenzhen, Guangdong, China
Employee Strength – 14000
Industry –Videography, Photography, Consumer Electronics, and Technology
Recent Achievements – Presented its first industrial drone with hybrid cameras- Matrice 300RTK with advanced artificial intelligence capabilities.
AI Company – Artificial Intelligence at DJI
DJI has been the long market leader for manufacturing some truly fantastic devices with AI capabilities. DJI Phantom 4 is its new drone that uses depth maps and machine learning to intelligently avoid objects and track people. DJI has partnered with Microsoft to combine drones with AI so that they can be used for diverse industrial applications. In fact the pro version i.e. DJI’s Phantom 4 Pro drone uses artificial intelligence to spot fruits by training a new machine learning model. The drone shoots a 4K Ultra HD footage and then Microsoft’s AI is used to identify what the drone is seeing.
A Look to the Artificial Intelligence Future
A study by Accenture estimates that artificial intelligence has the power to add up to 15% of the current gross value ( approximately $957 billion) to India’s economy by the end of 2035. As artificial intelligence becomes the magic wand to wave all the tech problems, the market for AI-linked technical skills is overheated in India. No matter if you are the CEO of a company or just a graduate, having AI-linked skills and the ability to practically apply those skills to real world problems will be crucial in the years to come.
Artificial intelligence jobs are becoming the career of the future making deeper inroads into the Indian IT industry with salaries ranging as high as 40 to 50 lakhs per annum. Now is the time to make a career transition into AI and become an integral part of the revolution in making the world a better place. Springboard India can help you build a promising artificial intelligence career with its comprehensive, project-driven, and 1:1 mentor-led Artificial Intelligence and Machine Learning program that comes with a job-guarantee.