Top AI Trends of 2022
I see so many Tech and A.I. trends on the horizon.
Artificial intelligence is set to increasingly scale in healthcare, drug discovery and dozens of other verticals. In the third quarter of 2021, investors poured $17.9 billion into global AI startups, CB Insights reports.
If you want to support my initiative to write about Artificial Intelligence, please consider becoming a premium subscriber to help fund my writing & research. I cannot do it without your help.
I remember when listicles used to mean something on the internet, and as an AI enthusiast let’s curate some of the chatter on this topic for the benefit of the end of year celebration.
Gartner’s Impact Radar is amusing:
Industrial AI and AI-on-5G internet of things applications is becoming more mainstream in 2022. Think about it, while we aim for the Metaverse, the way we are upgrading physical space is also impressive.
AI-on-5G combined computing infrastructure provides a high-performance and secure connectivity fabric to integrate sensors, computing platforms and AI applications — whether in the field, on premises or in the cloud.
Industry 4.0 like new automation and robotic systems
AI-on5G is generally associated with ultra-low latency in non-wired environments, guaranteed quality-of-service and improved security.
The Convergence of AI and industrial IoT solutions and the evolution of Edge AI makes this all possible and easier to implement.
2. Generative A.I.
Generative AI, or algorithms that assess existing data, such as text, audio or visual files, recognize the underlying pattern of that data and then replicate the pattern to generate similar content is becoming improving at scale.
As input data for models changes and as business outcomes change, the models themselves need adjusting. Lack of maintenance can cause the AI algorithms to eventually lose value.
You may have heard of this from all the various headlines in recent months.
There are various techniques such as:
1. Generative adversarial networks (GANs):
GANs are two neural networks: a generator and a discriminator that pit against each other to find equilibrium between the two networks:
The generator network is responsible for generating new data or content resembling the source data.
The discriminator network is in charge of differentiating between the source and the generated data in order to recognize what is closer to the original data.
They are trained to understand the language or image, learn some classification tasks and generate texts or images from massive datasets.
3. Variational auto-encoders:
The encoder encodes the input into compressed code while the decoder reproduces the initial information from this code.
If chosen and trained correctly, this compressed representation stores the input data distribution in a much smaller dimensional representation.
3. Arrival of the Augmented Human-AI Hybrid Workforce
While WFM is the new normal in work, the future of work is being paired more with AI in an augmented environment. Any repetitive tasks you do could and will be automated.
Whether you work in HR, admin, marketing, sales or engineering increasingly AI/ML tools are enabling you to work better and be more productive. This is also just a normal part of the future of work.
For instance, AI/ML technology is widely used in fields of knowledge, such as law or medicine, to peruse ever-increasing amounts of data and find the right information for a specific task. So many white collar jobs have significant augmentation possible to lead to even more productive work and allow people to do what they are intrinsically good at.
In every profession, there will be the availability of smart tools that are AI-driven, helping individuals in that profession to work efficiently. This is generally referred to as the augmented workforce or Human-AI hybrid work.
4. Cloud and Edge Management in IT
While edge computing is rapidly becoming a must-have for many businesses, deployments remain in the early stages. Cloud and Edge native business processes will become more dominant in IT and more ubiquitous in the business world.
Some believe AI management will become the responsibility of IT departments. Here’s an example: to address the edge computing challenges related to manageability, security and scale, IT departments will turn to cloud-native technology.
Kubernetes for instance is a platform for containerized microservices, has emerged as the leading tool for managing edge AI applications on a massive scale.
IT departments that already use Kubernetes in the cloud can transfer their experience to build their own cloud-native management solutions for the edge. Greater adoption in third party and related services are anticipated.
5. A.I. in Cybersecurity
Cybersecurity, fraud, identity theft and many other crimes are becoming more and more common in an era of Web3 and related legacy services not catching up to the criminals.
Cybersecurity like Climate change, wealth inequality or global conflict of the geopolitical national rivalry variety is a global risk that is increasing.
As we see more and more involvement of machines in every facet of our lives, there is a potential risk of cybercrime, and it continues to be a problem. Killer drones, self-driving cars, automated energy systems, it’s a cybersecurity problem of many angles.
As you can guess the cyberattack surface in modern enterprise environments is massive, and it’s continuing to grow rapidly. This means that analyzing and improving an organization’s cybersecurity posture needs more than mere human intervention. AI presents many advantages and applications in a variety of areas, cybersecurity being one of them.
Breach Risk protection
Service downtime protection
The role of AI in cybersecurity needs to increase in an automated way. 69% of organizations think AI is necessary to respond to cyberattacks but the field needs upgrading in the 2022 to 2032 time period.
6. Larger and Better Language Models
What will GPT-4 of OpenMind be able to do? Will the Beijing Academy of Artificial Intelligence (BAAI) be bale to keep pace? 2022 will answer a lot of the question of how bigger and better language models will be able to create new jobs, new apps and new business models — new kinds of technology startups that will change the internet and help us organize content in the Metaverse.
Larger AI models may allow for new possibilities with what A.I. will be able to do and how it will be able to learn. AI and machine learning models uses huge volumes of data and these models will continue to expand and draw on even greater data sets to make increasingly accurate decisions.
While the continued evolution of OpenAI’s large-scale generative pre-trained transformer (GPT) models get the fashionable headlines, what DeepMind, Microsoft Research and others do will bear watching. There are dozens of new startups around highly evolved large AI language models.
Where will it lead us in 2022?
Some analysts believe or reveal that GPT-4 may contain roughly 100 trillion parameters, making it 500 times larger than GPT-3. We can speculate that this development is literally a bigger step closer to creating machines that can develop language and engage in conversations that are indistinguishable from those of a human.
7. A.I. Applications in the Metaverse
What might be the possible applications of A.I. in the Metaverse and more immersive work and social environments in virtual reality and related to the evolution of a consumer brain-computer-interface (BCI)? How will mobile phones eventually be disrupted?
Metaverse is a terminology coined for an environment, a digital environment to be more specific, where multiple users can work and play together. If we play today on platforms with dumb algorithms and recommendation engines, what indeed will be the AI of tomorrow that helps us navigate and monitors our future work, social and dating life in the virtual world?
New kinds of apps, smarter digital agents, deepfake humans (that are actually bots), all await us in the future of the internet that seems to be Metaverse products.
8. Democratization of and Accessible AI — Low/No Code A.I.
Will A.I. ever truly be democratized? Will some of the wealth generation of Billionaires in a more automated world be distributed to the rest of us? It’s not crypto that saves the planet in this sense, it’s low/no code A.I.
People will be able to start new businesses without needing a team of expensive engineers in the future or having very specialized skills. While the demand is high today for A.I. engineers, imagine a different world. A world where A.I. can code itself. A.I. will eventually be able to alter its own code, and in 2022 I believe we’ll make breakthroughs in that direction.
One of the major challenges that organizations are facing today is the dearth of skilled AI engineers, who can develop the required tools and algorithms. With the advent of no-code or low code solutions, this challenge can be addressed by providing simple and intuitive interfaces, that can be used to create complex systems on Artificial Intelligence.
As we hasten AI adoption in business and upgrade A.I. processes and with coders working with AI-human systems, the way we create products using software engineering will fundamentally change and be more accessible to all thereby distributing some of its value in a more decentralized way.
Do you enjoy my content? Please consider being a Premium subscriber so I I can grow the Newsletter and my related Newsletters you can see here.
A related Newsletter I’ve started is called Data Science Learning center.
If this Newsletter is well received I will continue this list, you can give it a like (by tapping the heart icon) or a comment if you’d be interested in reading more of my analysis. Thanks for reading!