Top Machine Learning Trends & the Future of AI
What is the future of machine learning? Are we headed for an AI winter or an AI renaissance? The commercialization of AI is big business, but a lot of it is hype and gimmicks.
The Covid-19 pandemic has impacted many aspects of how we do business. But it hasn’t diminished the impact and rising importance artificial intelligence is having on our lives, our industries and the emerging world of Big Data.
While AI will increasingly be monitoring and refining business processes, it will also disrupt many industries leading to the automation of jobs. It brings convenience to consumers and some aspects of capitalism. It also presents the challenges of a more advanced surveillance capitalism.
Research shows that 77% of the devices that we currently use have AI built into them. The Internet of Things (IoT) isn’t just a buzzword anymore. Smart cities like we are seeing in China bring a cashless society that tracks our every move.
Even during the pandemic of 2020 we have seen AI both fail and succeed in small ways. Thermal cameras and similar technologies are being used to read temperatures before individuals enter busy places like public transportation systems, government buildings, and other important areas.
A more automated logistics is also appearing in an era when digital transformation moves into the brick-and-mortar reality of our lives. Robots are being deployed to implement “contactless delivery” for isolated individuals, helping medical staff ensure that key areas stay disinfected and safe for use.
Major chip manufacturers including Intel, Nvidia, AMD and ARM aim to produce AI-powered chips to speed up the operations of applications that run on AI. Major technology companies are also creating their own AI geared chips now. Software engineering is quickly evolving to keep up with new trends in AI and machine learning.
According to Gartner’s 2019 CIO Agenda survey, the percentage of organizations adopting AI jumped from four to 14% between 2018 and 2019. If you consider the big picture, adoption of AI is still very immature. By 2023, Gartner predicts that 40% of infrastructure and operations teams in large enterprises will use AI-augmented automation, resulting in higher productivity.
In the 2020s, the health and financial sectors will undergo rapid improvements due to the impact of AI and machine learning at the intersection of many industries and new emerging businesses (startups).
AI and machine learning skills are also seeing a huge demand for the GenZ and Alpha cohorts. Meanwhile, revenue generated by AI hardware, software and services is expected to reach $156.5 billion worldwide this year, according to market researcher IDC, up 12.3 percent from 2019.
The evolution of super computers, quantum computing and new kinds of research labs like DeepMind, OpenAI and others in China will create new synergies and new possibilities for the emergence of real progress in machine learning and AI.
Ant Group’s IPO and Saudi siding with Alibaba will also create more funds for the AI developments led by China. Additionally, Softbank’s next vision fund is to be more AI specific and create new avenues for the acceleration of machine learning.
AI will skyrocket wealth inequality in the wild wild west of ML profiteering
As adoption of AI and machine learning is occurring in many industries and businesses, however, few are seeing its benefits. The economic benefits go to those AI-as-a-Service companies like Amazon, Microsoft, Alibaba, Google and Baidu.
AI will scale wealth inequality in the 2020 to 2050 period like nothing else ever invented. This is a very dangerous aspect of AI’s progress that isn’t being talked about enough in BigTech. You won’t see it as a topic on SCMP or other Chinese propaganda sites or in Silicon Valley controlled outlets.
Most of the articles on AI are extremely bullish, full of hype and recycled material. To many, Machine Learning may be a new word, but it was first coined by Arthur Samuel in 1952. And since then, the constant evolution of Machine Learning has made it the go-to technology for many sectors.
Machine learning has become part of our lives, from smart cities to how new businesses scale their products and customer base. But AI and machine learning is not being regulated properly and we lack the legal framework to do so. This means it’s the wild wild west of machine learning profiteering in the 2020s.
AI undoubtedly remains a key trend when it comes to picking the technologies that will change how we live, work, and play in the near future. It’s not without its dangers: increasingly as BigTech abuses its monopoly dominance in our lives, the abuses of algorithms in our lives become more apparent. Algorithms are responsible for a toxic attention economy and $Trillions of dollars in lost productivity.
A toxic internet is the result of placing digital advertising at the center of the internet’s business models (Google and Facebook). It’s also profoundly reduced our mental health, increased civil unrest and made us more lonely. We need to think about the ethical impact of machine learning and AI on the world, not just its benefits.
Digital Advertising has also contributed to a pyramid winner-takes-all capitalism that has stunted innovation and technology startups in the U.S. since around 2010.
That the most talented software engineering and AI specialists go to the same firms is also responsible for lack of innovation in the U.S. It has created an innovation bottleneck in Silicon Valley that led it to being disrupted by China (in process).
Adoption of Machine Learning in the wider market requires higher level frameworks and advanced development tools. TensorFlow has led the charts among all the sources — Google, Github and Stackoverflow. This makes it a first choice for the developers. Keras is the second source.
Most framework people are looking out both on Google and Stackoverflow. TensorFlow is an end-to-end open source platform for machine learning. It has a comprehensive, flexible ecosystem of tools, libraries and community resources that let researchers push the state-of-the-art in ML and developers easily build and deploy ML powered applications.
Automation’s Effects: a significant decrease in jobs and businesses
AI and machine learning are key components – and major drivers – of hyper automation (along with other technologies like robot process automation tools). The danger of AI on job automation and the disruption of jobs is being under-reported by powerful lobbying groups, BigTech billionaires and dozens of studies that don’t come to good conclusions.
Retail, trucking and millions of lower level jobs indicate that automation will decrease the labor force significantly. We are already seeing this during the pandemic as many jobs in travel, hospitality, retail and small businesses have been lost.
While it’s easier to blame a pandemic than the rise of automation, Covid-19 fuels more automation as increased demand for efficiency becomes the new normal.
The pandemic has a greater chance of creating a machine learning renaissance in 2023 to 2035 than an AI winter scenario. This is primarily due to the economic and technological advances China will make during that period.
The new wave of small businesses will employ more AI in how they grow and serve their customers. Any business from Salesforce and Alibaba to Spotify that can help make this happen faster will do incredibly well.
Only about 53 percent of AI projects successfully make it from prototype to full production, according to Gartner research. As AI becomes smarter and more integrated in our R&D cycles, turnaround times will increase as we are seeing in the impact of machine learning on the Pharma R&D cycle. The potential of machine learning to innovate in healthcare is immense.
The Future of Machine Learning:
Machine Learning is on track to be worth around $9 billion globally by 2023. Areas to watch in machine learning include:
- Convolutional neural network (CNN)
- ML-based time series analysis
- Regulation of digital data
- Machine learning in voice assistance: according to the emarketer study in 2019, it was estimated that 111.8 million people in the US would use a voice assistant for various purposes.
- ML in automated Cyber-Security systems
- Lower costs of ML adoption in business analytics
- ML adoption in R&D. AI engineering incorporates elements of DataOps, ModelOps and DevOps and makes AI a part of the mainstream DevOps process
- ML in the smart city and the overlap evolution of IoT with various industries
- Evolution of global collaboration in IDEs. For example take Jupyter Notebook. Github considers it as a programming language but it’s more of an IDE (Integrated Development Environment)
- Encoded feature vectors. Anomaly detection based on autoencoders that run artificial neural networks using unsupervised learning algorithms
- AI becoming more autonomous in how ML itself evolves (e.g. better prediction models based on the optimization of Big Data)