The race for the first Super-Intelligence is on. But first things first. An exponential growth in AI applications and Data Science projects with thousands of dedicated teams has created a demand for a disciplined approach to creating autonomous systems that come with a built-in threat, which is no smaller than the possible extinction of mankind.
DATA CENTRIC AI
Companies should prepare for the highly likely scenario, in which competitiveness will be driven by the company´s AI´s capabilties. Soon we will see machines enter the arena of judgement and decision making, thereby producing new data to learn from. Whatever we do, be it nice or naugthy, will become "training data" for any learning machine as well.
AI algorithms will soon decide about win or loose, no matter what the question. The data-sets they have been trained on will determine the quality of such decisions. Getting the best data-sets possible is key to creating autonomous systems that some day may decide about your companies faith. Controlled Application Spaces give you just that.
The Race is on! Try not to be last one with a cutting edge AI.
Welcome to Lakeside Analytics - we offer consulting services in the area of ML/AI project management, conceptual design, analysis and preparation of data-sets as well as the enablement of Controlled Application Spaces.
Business Intelligence, Analytics and AI can be used for many things: to reduce effort and costs, improve speed to market, streamline logistic, drive design and innovation, reduce complexity and change relationships with players in the eco-system. The general idea is, that AI can be applied to improve various business processes: enhance or enable products development, support management in making better decisions, automate processes, provide insights to the market situation and propose strategic moves to position the organization better. There are also many more specific objectives that could be driving an AI intitiative: for example, a company might want to better understand the behaviours and decision making process of its customer and it could use machine learning to analyze highly granular data on the various transactional data about its customer. It might also be interested in analysis and predictions about customer loyalty, potential churn and counter the findings with appropriate measures. The company may also decide to use customer provided feedback as well as industry trends and innovation insights to improve product design and portfolio.
Some deployments seem to yield more value than others. Organizations may place a strong emphasis on a rather broad use case in their analytics and AI strategy, they may be very specific about its outcome or they could simply give it a try and see what the AI comes up with. Three very different approaches, with very different value propositions. This decision of which approach to take should largely be driven by business strategy.
Competitive Advantage through ML & AI - Don´t miss your Chance!
See examples on how various industries create value using highly advanced algorithms, well prepared data-sets and controlled application spaces.
Precision medicine, a technology still in its infancy, is an approach to treatment that uses information about an individual's medical history and genetic profile and relates it to the information of many others to find patterns that can help prevent, diagnose or treat a disease. "Precision medicine is one of the most trending subjects in basic and medical science today," sais Zeeshan Ahmed, an assistant professor of medicine at Rutgers Robert Wood Johnson Medical School. "Major reasons include its potential to provide predictive diagnostics and personalized treatment to variable known and rare disorders. However, until now, there has been very little effort exerted in organizing and understanding the many computing approaches to this field."
In the past five years, interest in applying artificial intelligence (AI) approaches in drug research and development (R&D) has surged. Driven by the expectation of accelerated timelines, reduced costs and the potential to reveal hidden insights from vast datasets, more than 150 companies with a focus on AI have raised funding in this period. And the number of financings and average amount raised soared in 2021. At the forefront of this field are companies harnessing AI approaches such as machine learning (ML) in small-molecule drug discovery, which account for the majority of financings backed by venture capital (VC) in recent years, as well as some initial public offerings (IPOs) for pioneers in the area.
The spectrum of approaches for the application of AI in small-molecule drug discovery can be illustrated by a selection of companies that have raised substantial funding (Table 1) and/or signed major deals with large pharma companies (Table 2) in the past two years. Some of these companies are mainly focused on a particular stage of the drug discovery pipeline, such as target identification or compound screening, while others are aiming to establish end-to-end platforms in which AI tools are at the core in each step.
see more: https://www.nature.com/articles/d43747-022-00104-7
Transporeon, the leading Transportation Management Platform, announced on August 22nd, 2022 the acquisition of Tracks, a Berlin-based start-up with the mission to decarbonise the transport industry. AI-based analytics and prediction tools that take carbon visibility to another level, facilitating the reduction in one’s carbon footprint through the targeted use of primary data. Tracks is a carbon visibility tool providing data solutions to monitor and manage carbon emissions across all transport modes. To do this, the company uses AI-based analytics and prediction tools to enable shippers, carriers and logistics service providers to collect and optimise emissions data at source.
see more: https://www.transporeon.com/en/company/press/transporeon-acquires-tracks
Argo AI, 412 Food Rescue, and Parkhurst Launch Autonomous Food Delivery. To help tackle food insecurity, leading autonomy products and services company Argo AI has partnered with 412 Food Rescue and Parkhurst Dining to launch an autonomous vehicle delivery service to transport surplus food to local nonprofits in Pittsburgh. The service will pair Argo autonomous vehicles with 412 Food Rescue partners to pick up and transport meals to nonprofits who offer food assistance and supplies in the region. With a team of more than 2,000 people across the U.S. and Europe, Argo designs its Argo Autonomy Platform and Solutions to support autonomous ridesharing and goods delivery to benefit communities around the world.
more about this: https://www.argo.ai/company-news/argo-ai-412-food-rescue-and-parkhurst-launch-autonomous-food-delivery/
When it comes to geospatial and mapping data and how they are leveraged by organizations, satellites continue to play a critical role when it comes to sourcing raw information. Getting that raw data into a state that can be usable by enterprises, however, is a different story. As a German company, LiveEO is one of a small but growing group of startups in Europe capitalizing on increasing interest in space among investors in recent years. LiveEO’s platform addresses a specific gap between space tech and enterprise data. Satellites are collectively producing more data about our world than ever before, covering not just physical objects in the most minute detail, but thermal progressions, how systems are moving and more.
read more: https://techcrunch.com/2022/08/04/armed-with-19-5m-liveeo-plots-a-big-data-course-between-satellite-geospatial-information-and-industry
Automated interpretation of industrial X-ray images. Digital detectors and highly automated inspections systems allow manufacturers to significantly optimize their quality control and assurance processes. Due to the higher throughput of parts, the evaluation and interpretation of X-ray images is starting to become a costly bottleneck. Advanced technologies like Automated Defect Recognition (ADR) and Artificial Intelligence (AI) have the potential to significantly reduce the required time per part. Depending on the inspection standard and requirements the algorithms can be implemented in an as an assistance to the operator or fully automated.. PFW Aerospace GmbH has implemented our Artificial Intelligence (AI) based ADR prototype in a production environment as an assistance tool to increase the throughput and process safety of X-ray image interpretation.
more about this: https://visiconsult.de/smart-inspection/
As millions of people transitioned to remote work during the pandemic, criminals took advantage of the increased volume of online activity. Scams proliferated, according to the FBI. Human traffickers and drug cartels, long seeking ways to circumvent regulators and law enforcement, refined their money laundering through online marketplaces and cryptocurrencies that afforded anonymity while connecting buyers and sellers, the U.S. Government Accountability Office reported. To adapt to these threats, regulators and law enforcement authorities began to turn to artificial intelligence and machine learning to battle bad actors. These tools help officials spot trends and set rules to help banks and other institutions identify and report suspicious transactions.
Meanwhile, the Securities and Exchange Commission proposed rules to crack down on financial cybercrimes. Even if government agencies and financial institutions could hire everyone they needed tomorrow, they would still need AI and machine learning to manage the deluge of cybercrime that is now routine. The massive scale of online activity and the scope of the cyberthreat today is too large for teams of people of any size to handle, even in the public sector. AI and machine learning can take on that work in three areas:
- AI helps officials in the discovery process, finding so-called “unknown unknowns,” or problems that remain hidden from regulators, as well as patterns in data across multiple firms and systems that point to emerging threats and new behavior patterns.
- Ninety-five percent of suspicious activity reports are false positives. Banks are inundating the U.S. Treasury’s Financial Crimes Enforcement Network with irrelevant SARs, wasting resources that regulators and financial institutions could put to better use in finding real crime. Using broader data sets to capture all known risks and reduce alerts, AI can cut down on false positives by more than 75 percent, discovering crime more effectively so resources can be refocused on other, more genuine threats.
- AI and machine learning can help law enforcement single out the worst bad actors and gather sufficient evidence against them. Identifying unknown relationships and linkages between entities and people, spotting unusual changes in behaviors across entities, digital tracking of activities, and other investigative technologies help law enforcement pinpoint the most mysterious and elusive bad actors perpetrating the most damaging crimes. Within mountains of data, AI can recognize the genuinely suspicious patterns and inconsistencies that are the telltale signs of drug and human trafficking organizations’ money laundering, for instance.
read more: https://www.c4isrnet.com/opinions/2022/08/18/federal-government-using-artificial-intelligence-to-fight-cybercrime/
UK Government Policy paper - Establishing a pro-innovation approach to regulating AI - Across the world, AI is unlocking enormous opportunities, and the UK is at the forefront of these developments.
read more: https://www.gov.uk/government/publications/establishing-a-pro-innovation-approach-to-regulating-ai/establishing-a-pro-innovation-approach-to-regulating-ai-policy-statement
In the National AI Strategy (UK), the government set out an ambitious ten-year plan for the UK to remain a global AI superpower. The UK is already a leader in many aspects of AI, with a thriving ecosystem and a strong track record of innovation. But there is more to do to harness the enormous economic and societal benefits of AI while also addressing the complex challenges it presents. Establishing clear, innovation-friendly and flexible approaches to regulating AI will be core to achieving our ambition to unleash growth and innovation while safeguarding our fundamental values and keeping people safe and secure. Our approach will drive business confidence, promote investment, boost public trust and ultimately drive productivity across the economy.
get the insights: https://www.gov.uk/government/publications/national-ai-strategy
Similiarly the EU is publishing numerous documents on the definition of AI and the constraints/differences of its use in public services.
see more: https://publications.jrc.ec.europa.eu/repository/handle/JRC129301
WHERE TO START WHEN GETTING INTO AI
Getting into AI requires a project, data and some champions around the organization. From a project management perspective, many traditional aspects need to be considered as well (regardless if you go waterfall, agile or something new): you want to clearly understand the business objective, you need to identify the gap (starting where you are right now, going to somewhere new) and its consequences, you need to get your data and data-sets right (streamlined, integrated, cleaned, prepared), you want to have proper controls and exit strategies in place, you want to create a change movement that aligns your organization to whats coming and last-but-not-least you want to make sure that you have skilled, available and motivated experts on your team. This may not sound like a milestone plan, but here´s your first lesson learned: The AI will tell you when its done!
IT's THE DATA, STUPID
If you look around you may spot a relatively high number of cases, where the results or successes in AI developments are leave a lot of room for improvement, the put it mildly. Clearly there are some projects that have a lot of success, but there are two or three times as many, that aren´t getting anywhere. What´s the difference you may ask. Interestingly, the difference is neither the tools and technologies used (they all us pretty much the same) and it is also not the skills, experience and capabilties of the team-members (they are all experts). So what is it? Bottomline: what makes an AI project work is the data, not the code, the model or the desired functionality. There is already wide agreement that projects need to move from a model-centric to a data-centric approach. But this move also needs to take place in the heads of those who manage the project, especially the stakeholders and customers (internal or external). If you agree that the benefits of an AI is the insights that it gains from its data-sets, then you are already one step ahead.
ML & AI PROJECTS ARE NO ISLANDS
What sounds rather discrete will eventually turn out to be extremely pervasive and all-encompassing. Except for when the project is closed-up and focussing on something that has no ties into improving the way the organization works, the need for high and tight integration is eminent. Large Business Intelligence projects have over and over shown how important a well planned data strategy (along with consideration of its heterogeneous systems, bespoke interfaces and many satelite projects) is. Without all hands on deck (technological, resource-wise and political) the project is doomed to fail. A BI project may still come to life in a reduced and isolated kind of application, but that´s limiting its potential. The difference is the horizon of data that is needed. While a BI project reports on the status quo, the Machine Learning and Artificial Intelligence objective is different: its about using as much data as possible, in a cleaned-up and smoothly prepared way to predict, judge and eventually derive decisions from it. Not only is the value of such projects greater, when the data is complete, comprehensive and conclusive, but also trust in its findings increases when each and every aspect is included. Any algorithm that is targeted for autonomous decision making, needs to create confidence and a sense of security, even though, eventually, it may mature beyond what is humanly comprehensible.
WHAT WOULD AN AI DO?
The question that it all comes down to is "what would an AI do?" If we accept that, like in the case of the algorithm that beat human players at "GO", at one point the algorithm reaches a maturity that will be incomprehensible by us humans, then ultimately we need to think about what may happen, once the AI if ahead of us. Will we be able to control anything at all? Of course, many applications restrict the AI to just a few functions. But some might not. Lakeside Analytics offers a 2 hour Seminar leading you through a number of already existing AI applications, with a focus on the perspectives that each area offers as well as a critical look at what it takes to stay in control. Contact us at firstname.lastname@example.org.
Get it right the first time!
Don´t wait until it is too late. Contact us today!