An article in the New Scientist caught my eye this morning. It announced that EEG technology and AI have been combined to create a non-invasive thought-to-text device. The initial 40% accuracy rate is already increasing to 60% in the latest studies – so this has great potential – despite the ground still to be covered.
I have been an avid follower of developments in Brain-To-Machine (BMI) interfaces since 2012, when an AKQA project for Deutsche Telekom introduced me to Harvard Braingate, and the visionary worlds of Ray Kurzweil and Aubrey de Gray.
I was deeply excited by the AlterEgo experiments at MIT in 2018, but since then progress seemed to have stalled. However with the advent of generative AI, I was confident that progress would accelerate rapidly, as generative AI is an accelerant for almost everything.
During the course of 2023, I have speculated a few times with ChatGPT on what the development path for BMI consumer-grade devices might look like. So prompted by today’s article, we had another session to revisit our assumptions and predictions.
Here is the result of our collaboration. I am super excited for what the next 12 months might bring to the game-changing world of BMI technology!
We started by comparing the relative advantages of EEG and sub-vocalisation techniques, since they work in really different ways. Then we explored a combination of the two techniques in one device – and speculated about likely time-to-market and key use cases for the first consumer-grade device.
Advantages of a Hybrid BMI:
- Enhanced Accuracy. By amalgamating EEG’s broad neural pattern reading with the precision of sub-vocalization, we could achieve unprecedented accuracy in capturing user intent.
- Error Correction. This hybrid approach could offer a fail-safe mechanism, ensuring consistent performance even under challenging conditions.
- Versatility. A combined device would cater to a wider spectrum of applications, from complex command execution to broad cognitive assessments.
- Accessibility. Tailoring to diverse user needs, especially those with disabilities, could be a hallmark of this technology, making digital interaction more inclusive.
Potential Use Cases:
- Complex Task Management. Imagine managing intricate software or playing strategic games using a blend of broad-stroke decisions and detailed commands, all through thought.
- Assistive Communication. For individuals with speech or motor impairments, this technology could be a beacon of hope, offering new ways to interact with the world.
- Educational Tools. The ability to gauge a student’s overall engagement and specific responses could revolutionize the educational experience.
- Personalized AI Interactions. This hybrid interface could lead to more nuanced interactions with AI, understanding not just commands but the context and emotional undertones.
Path to market:
Predicting the first form of Brain-to-Machine Interface (BMI) device to hit the market involves considering several factors, including current technological advancements, market readiness, user acceptability, and regulatory approvals. Based on these considerations, here are a few hypotheses – and a prediction for how the technology and product landscape might evolve.
Hypotheses for Market Readiness:
- Technological Maturity. Technologies that have been in development longer and have more established research, like EEG-based devices, may reach the market sooner.
- User Accessibility and Comfort. Devices that are non-invasive and user-friendly are more likely to be accepted quickly by consumers.
- Cost and Manufacturing. Technologies that are less expensive to produce and easier to manufacture at scale might reach the market first.
- Regulatory Approvals. Devices that pose fewer ethical and privacy concerns are likely to navigate regulatory pathways more easily.
- Market Demand and Use Cases. Technologies with clear, compelling use cases, especially in lucrative markets like healthcare, education, and personal productivity, might be prioritized.
Prediction: Given these hypotheses, it’s likely that an EEG-based device could be the first to hit the market. These devices have been around for a while, have a relatively established body of research, and can be designed to be non-invasive, which is a significant advantage in terms of user acceptability and regulatory approval.
As for the timeline, considering the current pace of technological advancements and the increasing interest in BMI technologies:
- Short-term (Next 2–3 Years). We might see more advanced prototypes or limited-use cases in specialized fields like healthcare or research.
- Mid-term (3–5 Years). Consumer-grade EEG-based devices, possibly in the form of wearable technology like headbands or caps, could become available. These devices might initially offer basic functions like mood assessment, focus enhancement, or simple command execution for interfacing with digital devices.
- Long-term (5–10 Years). More sophisticated versions, potentially integrating sub-vocalization technologies, could emerge, offering more nuanced control and wider applications.
It’s important to note that these predictions are based on current trends and could be influenced by breakthroughs in technology, changes in market dynamics, or regulatory shifts. The field of BMI is rapidly evolving, and unexpected advancements could accelerate the timeline.