Roundup of 2018 AI Predictions

“I can’t change my past, or predict my future. But I can shape my present.” ― Armin Houman

It’s prediction season, so similar to last year, I have taken some time to read through the top predictions for the year. No need to mention that forecasting is not easy, and you have to take these with a big pinch of salt, but it’s still interesting to extend an imaginary line from an existing trend and see where it leads.

In this post I decided to focus solely on AI prediction, not so much the ‘eye rolling’ apocalyptic ones, but the more practical view on how will AI progress next year.

The general trend and predictions for this year

  1. Ethics takes the centre stage in AI development and public discourse. More diversity will be included in training sets to reduce biases.
  2. Cameras will be everywhere leveraging machine learning to parse images in real time, enabling machines to see.
  3. Vertical applications of AI will create new business opportunities in many fields/professions — legal, accounting, health, wealth management etc.
  4. The large companies will continue to work on their own hardware chips to improve performance of AI.
  5. AI will continue to enhance human performance, automation and improvements in robotics will continue to increase productivity and performance.

I liked the Thomson Reuters’ guide of AI predictions:

“To move beyond the hype and look to the immediate future, we asked 10 Thomson Reuters technologists and innovators to make their AI predictions. for the year ahead”

  1. AI brings a new set of rules to knowledge work (“In the long term, our objective is to build personal digital assistants for knowledge workers”)
  2. Newsrooms embrace AI
    (“Imagine market reports that were written on demand and not just when the market closed. These reports could be more than just a simple recap of market performance, but a comparison of a how a reader’s portfolio performed against the broader market, as well as key reasons why”)
  3. Deep learning goes mainstream
    (“By teaching machines to understand unstructured data and filter out the most relevant, they lessen the burden of time-intensive processing work and allow experts to spend more time on high-value work like talking to clients”)
  4. Machine bias and algorithmic diversity come into view
    We’ll see an awakening to the fact that we need AI programming (and programmers) that cuts across age, race, socioeconomics, and gender to truly capture all of the inputs that go into making smarter, more rational, and unbiased decisions

See the full list at:

An thorough round up from CognitionX on predictions in various categories for AI: Future of work, Transportation, retail, investing healthcare, etc.

Some of my favourite predictions:

  1. According to recent survey findings from Gartner reveal that almost 60% of organisations surveyed have yet to take advantage of the benefits of AI. 2018 will be the year when more and more enterprises will take the plunge and go from Lab to Live.
  2. Safwan Halabi (medical director of Radiology Informatics, Stanford Children’s Health, Lucile Packard Children’s Hospital): we will begin to see AI-enabled tools translate from the research lab to the radiologist workstation and ultimately the patient bedside
  3. Georges Nahon, CEO, Orange Silicon Valley: the face will be the new credit card, the new driver’s license, and the new barcode

See their full list here:

Azeem curates the Exponential View newsletter, an excellent recap of AI, ML and future tech. In the latest issue, Azeem shared his own 2018 predictions, some of them specifically about AI. Here are my favourites:

  1. International relations, the political economy and governance, will desperately need new design patterns as we enter a new phase of the digital revolution.
    (“National AI strategies will emerge from more countries. The result? More grounds for co-operation and more reason to argue about intellectual property, privacy, data and license to operate”)
  2. The AI software stack will continue to diverge from traditional software
    (“Novel interface mechanisms. One will be voice, both as an input and as the output. The second will be images. Embedded cameras are providing large-scale inputs to machine learning systems: computers can now see. (One example will be the growth of affective computing applications.)”)
  3. Artificial intelligence will be the technology investment priority for large firms.
    (“After years of prototypes, automation technologies and AI software now dominate the CIOs agenda. They will invest and invest big”)
  4. We will increasingly demonstrate how AI is augmenting human capabilities.
    (“We will see more evidence for the tangible benefits the superpowers AI tools can give us individually, and we’ll increasingly witness the power of the AI-augmented human”)
  5. The discussion on how AI will impact employment will shift from solely focusing on the elimination of jobs to how best to help workers accommodate the inevitable change.
    (“We will also make more progress in understanding the commons questions of trust, fairness and justice in algorithmic systems. Sensible boards, prompted by legislators, regulators and activists, will make ethical AI a top-table issue”)

See Azeem’s full list here:

There are few people with the vantage point that Mustafa Suleyman, co-founder of Deepmind, have on AI. In a Wired column, he predicts that ethics is going to take the spotlight in the study of artificial intelligence as we leverage AI to tackle the world’s biggest problems from climate change to feeding a growing population.

They could either help overcome economic inequality or they could worsen it if the benefits are not distributed widely. They could shine a light on damaging human biases and help society address them, or entrench patterns of discrimination and perpetuate them.

This is why I predict the study of the ethics, safety and societal impact of AI is going to become one of the most pressing areas of enquiry over the coming year.

What’s certain, is that it won’t be a quick win:

Progress in this area also requires the creation of new mechanisms for decision-making and voicing that include the public directly. This would be a radical change for a sector that has often preferred to resolve problems unilaterally — or leave others to deal with them.

Fast Company asked VCs from eight firms for the most important tech trends in 2018. Here are their AI predictions:

  • The Human Side Of Algorithms — What are the investment opportunities in human impact?…Not removing people from the equation, but giving people superpowers. (Hailey Barna, First Round Capital)
  • Specialized Artificial Intelligence And Machine Learning — companies focused on single-purpose uses of AI will create new business opportunities. Automation will provide a massive multiplier effect by being able to do the small tasks that thousands of people could do–like look for patterns in images. (Ben Evans, A16Z)

Alexis Madrigal at the Atlantic, has 8 overly confident mostly pessimistic predictions about tech in 2018 — when it comes to AI he said

  • Fake News is our new reality —

“New artificial-intelligence techniques portend an even wilder reality, where media is even less trustworthy. Neural networks, a particular form of machine learning, can generate new audio from archival audio, making it possible to put words in the mouths of public figures. They can also generate video scenes. This is our future. People who want to really know what’s going on will have to rely on gatekeepers, whether that’s the mainstream media, analysts at research firms, or other information services.”

  • Cameras will be a part of ever more devices, even if they don’t have screens.

“Cameras have become so cheap that they are ubiquitous. They exist in all kinds of products, from cars to doorbells to baby monitors. In the past, cameras existed to display things to humans. But with the rise of machine-learning techniques that allow computers to parse images, the biggest users of cameras will be machines. And that means manufacturers will build cameras into ever more devices, so that they can exhibit low-level intelligence about their surroundings. While there will be privacy concerns, some of them will be alleviated by the devices doing more processing locally without sending information to the cloud, thanks to new chips that can run the artificial-intelligence software needed for image recognition.”

Some other interesting AI predictions and lists:

  1. Hackernoon’s Machine Learning and AI trends for 2018 — what to expect?
    Good overview by sector (health, fintech, robotics…)
  2. Forbes 51 AI predictions for 2018 — Gil Press asked 51 CEOs what 2018 holds for AI. They most “envision it becoming more practical and useful, automating some jobs and augmenting many others, combining machine learning and big data for fresh insights, with chatbots proliferating in the enterprise.”
  3. The Drum weighs on the implications of AI on brands and marketing — The AI pretenders will get caught out
  4. 10 predictions for deep learning in 2018 —
  5. AI in 2018: Experts predict what happens next —
  6. Massive Cyber Attacks, Autonomous Electric Cars And ‘Augmented’ Workforces: Israeli Industry Experts’ Predictions For The Next 5 Years —
  7. Accenture’s report “Artificial intelligence is the future of growth”
  8. Deloitte Global TMT Predictions —Artificial intelligence (AI) to support business growth as digital content elevates the entertainment industry
  9. AI In E-Commerce — Predictions For 2018.‬

Managing Partner at Remagine Ventures. Founder of Techbikers, Campus London and VC Cafe, proud Xoogler. On the boards of Chargifi, HourOne, Vault AI and EchoAR