Monday, October 23, 2023

Why Do You Need to Define AI?

Everyone is talking about AI, and nearly half of the digital/technological budgets for 2024 are aimed at AI. Because of this, everyone should care about what AI is and what kind of AI you want to meet your objectives. You need to know if something is AI or if some vendor or internal developer is just AI-washing what they have to offer. I think it is essential to understand what AI is today because AI is on a journey, and it's vital to know where it came from, what can be done with AI today, and where AI is headed. This way, organizations aren't stuck with pioneering without knowing they are on the edge or using traditional technology wrapped in an AI wrapper to leverage the AI movement. Or you are claiming AI victories by just putting window dressing on conventional technology. AI will be a critical player on your team soon, so to get to know AI early, I tried to identify authoritative definitions and put them in one place for you, including what one version of AI thinks. I also attempted to define a value-added definition of AI, leveraging my past business experience leveraging AI to get started.






General:

Artificial intelligence leverages computers and machines to mimic the problem-solving and decision-making capabilities of the human mind.

The Association for the Advancement of Artificial Intelligence (AAI): AI’s primary goal is to build an intelligent machine. The second goal is to find out about the nature of intelligence.

WIKI: Artificial intelligence (AI) – intelligence exhibited by machines or software. It is also the name of the scientific field which studies how to create computers and computer software that are capable of intelligent behavior.

Gartner defines artificial intelligence (AI) as applying advanced analysis and logic-based techniques, including machine learning (ML), to interpret events, support and automate decisions, and take actions. This definition is consistent with the current and emerging state of AI technologies and capabilities, and it acknowledges that AI now generally involves probabilistic analysis (combining probability and logic to assign a value to uncertainty).

Forrester defines “generative AI” as: “A set of technologies and techniques that leverage massive corpuses of data, including large language models, to generate new content (e.g., text, video, images, audio, code). Inputs may be natural language prompts or other non-code and non-traditional inputs.”

Investopedia

The simulation of human intelligence by software-coded heuristics

ChatGPT

AI, or Artificial Intelligence, refers to the simulation of human intelligence processes by machines, particularly computer systems. These processes include learning, reasoning, problem-solving, perception, and language understanding. AI technologies aim to enable computers and machines to perform tasks that typically require human intelligence, such as understanding natural language, recognizing patterns, making decisions, and adapting to new information.

Jim Sinur


"AI is the leverage of software and machines to add perception/intelligence to individuals, customer/constituent experiences, processes/tasks and devices to optimize balanced outcomes by interpreting patterns of interest, making highly informed decisions with speed, and taking appropriate proactive or reactive actions considering wide and deep implications all within the context of changing conditions and governance guardrails."

Net: Net:

It is essential to know what kind of AI you are buying. Machine intelligence or generative AI is often represented as the only and most advanced AI. Ensure you understand the AI you are buying or building as many technologies participate in the AI disciplines. It may mean understanding the multiple streams of AI contributing to a solution you may buy or build. It influences what problems you are trying to solve, your testing/debugging, and where it can go off the ranch and get into trouble in places you don’t anticipate. If you want a picture of where AI has come from or where it is likely to head, please click here. The Gartner AI Hype Cycle is another good resource. Better yet, you need to define what kind of AI you want to pursue in line with your business objectives and get ahead of AI’s projected core competencies and technologies for the future.

It would be best if you defined AI for you continuously as it evolves to the ideals declared in the general definitions available. AI is just a set of methods, techniques, and tools that will be used for good and bad outcomes. Remember, AI is not God, and nor should it be used by actors to create an artificial god. Humankind will eventually be supercharged with AI. As always, some will fly too close to the sun.


Additional Reading:


Wednesday, October 11, 2023

What Have Peeps Been Reading in the 3Q 2023?

 Topics of customer journeys and processes continue to gather interest, but surprisingly, the combination of IOT and process hit the top spot for interest. Of course, AI is peaking because of the momentum of generative AI. Managing new balances with collaboration, balancing new digital efforts with legacy maintenance, and creating the elusive management cockpit for management visibility also gathered interest. Sweden and Canada were the most active offshore countries. 

                                          Hot Topics for 3Q2023



                                                Third Quarter 2023 Offshore Activity  (non-US & China)           




Monday, October 2, 2023

Who is Afraid of AI?

With all the AI-related newsfeeds, stories, announcements, and tech giant personalities sharing their wisdom, it would be hard to avoid hearing about the "Big Bad AI" undercurrents. To answer this question honestly, I would admit to both fear and excitement. For the short term, the news is mostly good and helpful, but the fear of where AI might end up down the road scares us all. To sort this out, I tried to identify ten things that scare me about AI and ten things that encourage me about AI. Read about the three Eras of AI coming your way by clicking here.



My Top Ten Fears

AI Takes Over the World

At the worst, AI will become self-aware and use its powers against humankind. I'm not a big believer in this scenario. While AI will network with other AI forces to do good, it is more likely that bad actors will leverage AI for dark outcomes or power than AI coalescing to destroy humankind.

AI Lacks Ethics and Empathy

AI is great at doing tasks today informed by multiple data, information, and knowledge sources. As AI spreads, it will become more engrained in the decision-making processes at various levels, and decisions will likely be driven by hard science and logic rather than the feelings of people or the respect of ethical behavior.

AI is Used to Battle Security

AI will be used to fool individuals and organizations with deep fakes by computing through multiple security defenses. We see this emerging now, but AI can be used to battle security incursions. The arms race will only get more intense with AI supercharging the security wars.

AI Displaces Jobs & Skills

AI will take jobs away from people. It will start with menial or manual work, especially where danger is present and repetitive precision is needed. While AI will create new jobs that require new skills, people will be displaced until they find work AI is not great at, which tends towards creativity and careers that need deep people skills. The workforce will always have to be learning or chasing the last chair in a game of AI musical chairs.

AI Lacks Transparency & Explanation


AI and automation must rarely explain themselves or be completely transparent. AI must explain itself, at least after the fact, to govern and deliver fair treatment. Ideally, AI should ask before, but that takes time. Time is often the savings benefit that drives AI, so that post-audit trends will be vital.

AI Lacks Real Creativity

Yes, AI can copy creations of the past and even generate projects based on creative libraries of content, but will it be able to create new concepts that please the nature of human appreciation? There is much room here for AI to generate and have humans add or adjust, but the natural creativity lies in humans today.

AI is Used for Social Manipulation


AI can fake stories, create deep fake videos, and play impostors cleverly. In the hands of manipulators, AI can be leveraged to develop actions in humans who buy, vote, and act. While it can be used for good, like changing behaviors to benefit societies, AI can also steer us toward bad outcomes.

AI Invades Privacy

AI can listen everywhere simultaneously across various communication channels and existing data fabrics to expose information that individuals would not want available to the public or particular parties. It is scary for most folks and can be used to breach the trust and security of many relationships with individuals, businesses, and the government.

AI Ignites Economic & Geopolitical Competition

AI will be the fuel for competition in the fast-growing digital economies. The nations that harness AI will have a distinct advantage over those that do not. It will likely turn into an arms race of sorts.

AI Enables Laziness & Skills Atrophy

AI will show significant promise and results in assisting organizations and individuals. Specific skills will not be maintained, and people will want AI to do more for them. Some of the skills are mundane, so that might be good, but setting an entitlement attitude is not a great value to deliver from AI.

My Top Ten Encouragements

AI Enables Advanced Automation


Organizations are in love with Automation because of its positive effect on profits. AI will supercharge Automation with smarts, speed, and precision. Operationally, AI will be a big win. Low-level work will be eliminated, and AI will enrich and augment most jobs.

AI is Always On

AI never rests and is available 24/7 if the AI infrastructure and applications are running. People need rest, and AI does not. As AI progresses to higher-level skills, new work classes will inherit a new level of availability.

AI Provides Real-time Knowledge & Wisdom

AI is excellent at providing just-in-time data and integrated, summarized, and massaged information. In the first era of AI, there will be a big emphasis on machine learning, information aggregation, pattern recognition, and knowledge delivery. All of this will help workers and individuals progress in their desired outcomes.

AI Assists in Task Completion

Not only will AI deliver knowledge, but it will also help people make decisions and perform tasks of all kinds. People and bots will be supercharged with additional perceptions, projections, skills, and abilities to take on work over their current station.

AI Specialization & Focus Delivers Precision

AI is so precise on low-level tasks it outperforms most workers. In addition, even when 100% precision is not necessary, AI can give the best alternatives with the best accuracy.

AI Delivers Better & Faster Decisions

AI is so fast and looks across multiple contexts, and it's impressive. Guided AI can find the best alternatives given even conflicting goals within explicit constraints. The data is complete better than other approaches, and alternative algorithms have the best chance of being correct.

AI Boosts Economic Growth


AI increases the productivity of many resources, so profits and GDP will flourish worldwide. Resources will be less stressed, and there will be more free time to generate new products and services.

AI Provides Continuous Monitoring of Results

Since AI never sleeps, KPIs, goals, outcomes, and emergent trends can be watched. AI will encourage continuous feedback, and improvement can be baked into responses.

AI delivers Error Reduction.


AI does not make mistakes at the operational levels and allows tactical and strategic management to get the best results in a modeling way.

AI Can Complete Dangerous Tasks

Tasks that risk the safety of humans can be automated with AI, or humans can be assisted safely.

Net; Net:

After putting down my thoughts and thinking deeply about AI, I fear AI long-term. If the control of AI ends up in the hands of bad actors or AI becomes self-aware and disregards its guardrails or constraints, we are in for a wild ride. AI-driven wars where guardrails are removed or mismatched will also cause bad outcomes. AI likely becomes as powerful as nuclear weapons in the hands of both good and bad actors that mutually keep each other at bay. Right now, AI promises to improve our lives, and I see a rosy outlook for the near term. AI will contribute to individuals, groups, and organizations for sure. The long-term evolution and potential misuse keep me up at night. We must forge ahead to stay competitive, but AI must be monitored, governed, and balanced. There will be tremendous and sad stories ahead of us, so keep your eyes open and stay in a learning mode together. Long live AI, but don't take the oxygen out of the room for humans.