Exponential Technologies Underpinned by AI

Authored by Sameer Dhanrajani, Chief Strategy Officer, Fractal Analytics

Combining Exponential Technologies with AI 

Blockchain and AI 

In recent years, we have seen immense excitement among technology vendors and investors in the area of Blockchain. Estimates from Gartner suggest that blockchain applications will create $3.1 trillion in business value by 2030¹. Given its immutable and decentralized nature, blockchain will be invaluable in sectors such as manufacturing, defence and financial services – and we will see innovative use cases coming out of these domains. Over the past few years, several businesses that are mature on the technology adoption scale have incubated their first few blockchain experiments. With an underpinning of AI, blockchain can deliver much more substantial value. 

Blockchain and AI can come together powerfully, in use cases concerning fraud and security. We have seen numerous crypto frauds emerging in the past year that blockchain would do well to protect. By introducing anomaly detection in the blockchain environment, corporations would be able to stream real-time reports on the health of the overall blockchain, detect intrusions and provide diagnostic reports on how such attacks can be mitigated in the future. 

Cybersecurity and AI 

We live in an increasingly vulnerable world today. Our understanding of key competitive assets is rapidly changing and one of the most potent assets corporations have is their data. With many malafide actors looking to gain unauthorised access and exploit vulnerabilities in existing IT environments, cybersecurity is a subject that is seeing some massive interest. While cybersecurity is a remarkably advanced field already, by bringing AI into the equation, we will be able to further fortify our technology assets. 

Within the field of AI, anomaly detection and machine learning are specifically expected to hugely disrupt the field of cybersecurity. AI will empower security practitioners to identify intrusions and behaviour of malafide actors in real time using automated, always-on algorithms. AI could be employed to constantly survey a secure network for wrongful activity while addressing potential concerns before break-ins occur. 

AI can be quickly trained over a massive data set of cybersecurity, network, and even physical information. We are already seeing incumbent cybersecurity vendors who are rolling out AI-enabled solutions that learn to detect and block abnormal behaviour at an abstract level – even when this behaviour does not fit within a known pattern. 

Soon, we will see good AI and bad AI in the domain of cybersecurity compete against each other. There are genuine fears surrounding the fact that the next generation of attacks will be carried out by pieces of code designed to rapidly infiltrate a secure technology environment. Countering intrusions with ‘good AI’ will be crucial in undermining the impact these fast-paced attacks can have. 

AI to Fuel Edge Computing 

The next generation in the cloud computing paradigm is edge computing. Simply put, edge computing is a computing topology in which information processing, content collection and delivery, are placed closer to the endpoints. Edge computing paradigm is critical for advancing the cause of AI and there is a good rationale behind why such a synergy exists today between these two exponential technologies. Edge computing will soon become an enterprise mandate, given the value that it brings in reducing the latency in running AI algorithms and their response times. 

Adoption of edge computing is largely driven by the need to converge IoT and AI. By keeping processing power closer to the IoT endpoints as opposed to a centralized cloud server – it will provide a quantum leap in the joint value that these technologies are bringing together. Better yet, edge computing does not necessitate the creation of new architecture as cloud and edge computing complement each other. In most circumstances where AI and cloud come together, cloud services will be charged with centralized service execution, on both centralized servers, across distributed servers on-premises as well as on-the-edge devices.

According to Gartner, storage, computing and advanced AI and analytics capabilities will expand the capabilities of edge devices through 2028². Not too far into the future, we will see 40% of organizations' cloud deployments include an element of edge computing, and 25% of endpoint devices and systems will execute AI algorithms³.

AI Augmentation of DevOps 

Actual adoption of the DevOps framework across global enterprises has largely been patchy and slow, despite almost universal acceptance and appreciation among technologists. Now the reasons for that are numerous – from a distributed toolset to a clear lack of elite, specialized practitioners. However, we do see that with an underpinning of AI, we will see accelerated enablement of DevOps through increased process automation between software development and operations. 

The automation quotient in the DevOps process is on the rise and one of the reasons behind it is the increased adoption of AI-powered QA suites. AI is now rapidly intervening in the QA process – across areas such as unit testing, regression testing, functional testing and UAT, thus powering a better-orchestrated adoption of DevOps, underpinned by AI. Another such area where DevOps and AI come together is through DevSecOps – which combines both in the field of information security. Using a centralized logging architecture that records suspicious activities and threats alongside ML-based anomaly detection - can empower developers to pinpoint potential threats to their system more accurately while securing it for the future. 

Perhaps most importantly, AI will also break the cultural barriers that typically exist between developer and operations teams. One could argue that this has been the single most potent bottleneck that has hampered the success of DevOps. AI-enabled technology systems can enable DevOps teams and provide them with a single, unified view into system issues across a complex toolchain. 

Internet of Things Powered by AI 

IoT has hit mainstream status and enjoys high adoption across industries - as diverse as consumer goods and retail, to energy and utilities. This is due to the value it brings to multiple business processes across the board. IoT-enabled hardware devices are proliferating nearly every major walk of our personal lives as well. Devices from sensors, smart assistants and wearables are fast becoming a feature of everyday life, and AI plays a crucial role in enhancing their viability for use. 

IoT provides perhaps the most symbiotic relationship with AI. With an explosion in the adoption of IoT devices that are constantly monitoring for events, we are seeing an explosion of data being created by these devices. The quantum of data created by IoT devices is well beyond the human realm, and most certainly requires an underpinning of AI for extracting actionable insights. Using ML over IoT data can be extremely rewarding for companies – and can help them improve their forecasting abilities while deciphering real-time intelligence for business users to work on. 

Final Word: Convergence of AI, Blockchain, Cloud and IoT 

Could a future software stack comprise AI, Blockchain and IoT running on the cloud? There are high expectations of what might be possible should we be able to converge these technologies. IoT devices might become an interface where consumers and business stakeholders interact. Voice-enabled connected devices can augment the buying experience. In addition to this, AI frameworks such as speech recognition and NLP could be the translation layer between the connected devices and the technology that initiates the next action. Following on from there, Blockchain-like decentralized databases would be charged with managing contracts and transactions between various parties in the supply chain. Cloud and edge will, of course, be the mainstay for running these applications. 

In such a world which does not seem too far into the future, AI will remain as the key underpinning technology that will spur the development of other exponential technologies. Digital evangelists that are charged with the technology-led transformation of their enterprise should consider the wide-ranging impact of AI in the success of other exponential technologies. 

Sources: 

¹Based on the ‘Blockchain Business Value, Worldwide, 2017-2030’ forecast by Gartner, 02 March 2017 
²Based on the article titled ‘Gartner Identifies the Top 10 Strategic Technology Trends for 2019’, 15 October 2018 
³Based on the discussion paper titled ‘IDC FutureScape: Multiplied Innovation Takes Off, Powered by AI, Distributed Public Cloud, Microservices, Developer Population Explosion, Greater Specialization and Verticalization, and Scaling Trust’, 30 October 2018