Why might individuals and terrorist entities choose to experiment on their victims using the technology of discreet neural interfaces?

The development of neurotechnologies holds great promise for numerous applications, including the treatment of neurological disorders, the enhancement of cognitive abilities, and the creation of brain-computer interfaces.

The development of these technologies, including microelectronics, biotechnology, and quantum computing, has given nations and private industries a significant first-mover advantage, leading to a race to capitalize on the growing market for neurotechnologies. Consequently, the advancement of these technologies has become a major focus for many countries and private companies around the world. As the demand for neurotechnology grows, so does the potential for nefarious actors to use these technologies for malicious purposes

The United States is putting a great emphasis on the progress of neurotechnologies.. The United States' BRAIN Initiative has been recognized as one of the leading efforts in this field, aiming to improve our understanding of the brain's functions and the mechanisms that underlie different neurological disorders. The initiative's long-term goal is to develop new treatments and cures for conditions like Alzheimer's disease, Parkinson's disease, and epilepsy, among others.

In addition to the United States, China has also placed a significant emphasis on the development of neurotechnologies. The China Brain Project, which was launched in 2016, aims to create new technologies for both diagnosing and treating neurological disorders, as well as mimicking human intelligence and improving human-machine interfaces (Poo et al., 2016).

The project's framework involves a collaborative effort between researchers, engineers, and physicians, with the goal of developing new tools and techniques for studying the brain's functions and improving our understanding of various neurological conditions. The project is also working towards the development of brain-inspired artificial intelligence, which could have a wide range of applications in fields such as robotics, autonomous vehicles, and natural language processing.

The development of advanced technologies has been a subject of great interest and investment by numerous entities, including some organizations that use terrorist methods.Terrorist organizations or other groups may seek to hack the brains of scientists in order to gain access to brain data, sensitive information, or to cause harm. These attacks can take various forms, such as implanting false memories or altering a person's behavior or emotions. While these technologies have the potential to improve human life in countless ways, the problem arises when parties opt for simplistic solutions to address their complexity.

By doing so, they may inadvertently harm humans instead of ensuring their reliability and security. This is because hastily implemented solutions may fail to account for all potential consequences, leading to unintended negative outcomes. For example, an organization may choose to vandalize individuals as a means of achieving its objectives, but this could create a cycle of violence that ultimately harms everyone involved.

Conversely, organizations that take a methodical and deliberate approach to the adoption of new technologies are more likely to succeed in the long term. By investing in careful planning and testing, they can identify potential risks and address them proactively, ensuring that the technology functions as intended without causing unintended harm.

The key takeaway here is that technology adoption requires a balanced approach that prioritizes both speed and security. While it may be tempting to rush into new technologies, doing so without proper planning and consideration can result in significant harm. Therefore, it is essential for organizations to take a measured approach that balances the potential benefits of technology with the need to ensure human safety and security.

In the following passage, we elaborate on the reasons behind companies and organizations conducting experiments to obtain cognitive data from specific individuals. However, when the framework is unclear, and the entity conducting the experimentation is a terrorist organization, it could lead to brain hacking and illicit experimentation on people. This kind of experimentation is typically associated with torture and may result in violent behavior.

Companies and organizations are conducting experiments to gather brain data for various reasons. Namely, the technology of accessing brain data has revolutionized the development of artificial intelligence algorithms. These algorithms are capable of processing complex data and delivering accurate and fast results. With these algorithms, developers can create applications that can learn, understand, and act like humans, advancing the field of artificial intelligence. However, access to brain data has also dispelled many myths and may even go further.

In this regard, the availability of brain data has enabled the development of algorithms that support neural interfaces, with the aim of enhancing and assisting people with disabilities. However, it has also allowed malicious entities to steal sensitive data from victims and take control of their possessions and lives. Additionally, terrorist organizations can use this technology to carry out violent actions that further their goals, as well as to train their own AI models and develop their own neural interface softwares.

Namely, researchers are utilizing brain data to train their neural networks, specifically Convolutional Neural Networks to improve their accuracy in predicting, over and above machine learning techniques such as regression analysis.The idea of neural networks is to teach computers to think by giving it a very rudimentary understanding of cause and effect. This is done through a series of layers that work with mathematical functions such as convolution and elementwise multiplication. Convolutional Neural Networks have shown proficiency in object recognition tasks and predicting neural responses of the human brain. The brain data used in these studies was gathered through invasive methods from the inferior temporal cortex of a human test subject. Researchers have been able to achieve more accurate predictions and improved results; however, dealing with technical issues such as mode collapse and the difficulty of training is an unavoidable part of the process. Nevertheless, permanent evaluation remains an important factor. On the other hand, the ways in which a Convolutional Neural Network?can be used to predict a neural response with a non-invasive Brain-Computer Interface is still an open question

In a recent study by Wang et al. (2020), brain data collected through invasive methods was used to train Generative Adversarial Networks (GANs), leading to improved performance in computer vision, natural language processing, speech synthesis, and related areas of research. One of the most impressive outcomes of this research has been in image synthesis. The authors proposed an evaluation metric based on human perception gathered through a neural interface, which enabled them to train the GAN network and overcome the main challenges in GAN research: mode collapse, difficulty in training, and difficulty in evaluation. This metric made it simple to compare the difference between real and generated distributions, as accurately estimating the real distribution is an unachievable goal.

Another example of using brain data to develop novel algorithms with possible applications in many fields such as?vision and auditory tasks is the work of Lu et al.2022. Namely the researchers have developed a platform for neuromorphic computing and statistical inference called the brain digital twin?.?The Brain Digital Twin (DTB) is a new platform that enables neuromorphic computing and statistical inference. It consists of two components: first, a neuronal network is constructed and simulated at the scale of the human brain using high-performance computing and intensive biological data-driven structure. This leads to significant performance gains. Second, a proposed hierarchical mesoscale data assimilation method is used to fit brain resting-state experiment data and real-world functional experiment tasks. This method allows for the estimation of hyperparameters for over trillions of parameters, and a novel routing communication layout is used between 10,000 GPUs to implement the simulations.

The brain digital twin platform has been used for two experiments. The first, called the "dry" experiment, aimed to uncover information processing related to cognitive neuroscience in the entire human brain using either a visual or auditory stimulus. The second experiment focused on virtual deep brain stimulation (DBS) in medicine to explore underlying mechanisms and test DBS setups for individual brains. Two tasks were performed on the brain digital twin, one for visual processing and one for auditory processing.

With the brain digital twin?platform, scientists were able to simulate the human brain at the scale of up to 86 billion neurons. This digital twin brain is capable of mimicking certain aspects of its biological counterpart both in the resting state and in action. The use of a novel routing communication layout between 10,000 GPUs to implement simulations and a hierarchical mesoscale data assimilation method allowed for the achievement of over trillions of parameters from estimated hyperparameters.

Gaining a deeper understanding of how the brain works enables the development of more advanced technologies. This knowledge has led to the development of new technologies that can mimic or even exceed human capabilities in areas such as decision-making, learning, and problem-solving .Namely, Polykretis et al.2022 have taken inspiration from the structure and function of the brain to overcome limitations in neuromorphic computing, which have hindered its applicability to real-world robotic tasks. Specifically, they have utilized biomimicry to develop spiking neural networks that imitate the brain's asynchronous computation and topology. Rather than relying on traditional training methods, these networks incorporate knowledge of the brain's connectome associated with the targeted behavior. This approach has yielded promising results, as the behavior of the network emerges from the inherent structure of the spiking neural network, which mimics the topology of the associated brain areas. Furthermore, this method is biologically interpretable, allowing researchers to gain insights into how the system is operating, and requires no additional training to function. Additionally, the researchers have developed neuromorphic hardware to support the implementation of these algorithms, further increasing the potential for real-world applications.

Some firms are interested in hacking the brain and having advanced knowledge about it because it enables the development of sophisticated algorithms.?

The MIT researchers were able to develop "liquid" neural networks, thanks to studying the brain of small species?to closely approximate the interaction between their neurons and synapses . These networks are flexible and robust machine learning models that can learn on the job and adapt to changing conditions, making them suitable for safety-critical tasks such as driving and flying. However, as the number of neurons and synapses in these models increases, they become computationally expensive and require clunky computer programs to solve their underlying complicated math. Therefore, firms are looking for ways to unlock new types of fast and efficient artificial intelligence algorithms that can perform tasks involving time-series data, such as brain and heart monitoring, weather forecasting, and stock pricing. The new "closed-form continuous-time" (CfC) neural network introduced by Hasani et al.2022 is one such algorithm that is orders of magnitude faster and scalable, making it ideal for any task that involves getting insight into data over time. Overall, having advanced knowledge of how the brain works can provide firms with a competitive advantage in developing cutting-edge AI algorithms that can improve performance and decision-making in a range of industries.

Replicating human behavior has led to the development of robots that can engage in daily conversations with people. Researchers, including the authors of Hasumoto et al. (2020), have implemented the reactive chameleon effect, which involves mimicking the nonverbal behavior of conversation partners. This innovation has the potential to transform the service industry by providing customers with a personalized level of service that adapts to their individual behaviors and preferences. Such robots can cater to the needs of individual customers, revolutionizing the way that businesses interact with their clientele.

?There have been disturbing reports of certain terrorist groups attempting to hack into the brains of experts and scientists without their consent, in order to develop digital twins of their knowledge and behavior. Although this technique has been employed to gather information on complex systems, such as supply chain optimization, it is an unethical and invasive practice that violates individuals' privacy and intellectual property rights. Furthermore, this type of hacking is not limited to supply chain management and could extend to other domains, such as art and craftsmanship. In these fields, experts could have their creative processes and physical abilities studied and replicated in robots, potentially leading to the mass production of high-quality art or crafts.?However, the potential benefits of these developments should not overshadow the ethical implications of hacking experts' brains without their consent.

?The interest in understanding the brain is not limited to developing artificial intelligence models. While the development of sophisticated algorithms that can mimic the functions of the human brain has its own merits, there are many other potential applications of brain research.

One such application is the development of software that can facilitate the interface between the human brain and machines. This includes brain signal processing, which involves the analysis and interpretation of brain activity, as well as the development of tools that allow individuals to control devices using their thoughts, such as prosthetic limbs or robotic exoskeletons and the development of extended reality experiences, which can provide users with a more immersive and interactive experience by combining the physical and digital worlds. This requires a deep understanding of how the brain processes and interprets sensory information, as well as the ability to develop software and hardware that can seamlessly integrate the virtual and physical environments.

In summary,?The development of such neural technology can provide firms?with a range of benefits and open up new possibilities for innovation and progress.

Brain hacking can be done through invasive or non-invasive methods. However, as stated by?Zabir et al.2021?invasive methods have some advantages over non-invasive methods, which makes them preferable in some cases. Traditionally, invasive BCI refers to those interfaces that require surgery (e.g., Microelectrode Arrays or MEAS). The main advantage of invasive methods is that they provide higher spatial and temporal resolutions, enabling more precise data gathering and analysis.

Invasive BCI requires surgery, which can be risky and have significant health consequences. However, it provides direct access to the brain, allowing researchers to gather higher quality data and develop more sophisticated algorithms. Namely, they provide the spatial and temporal resolutions needed to productively encode and decode neural signals. This makes invasive methods suitable for certain types of BCI applications that require a high level of precision, such as clinical research and clinical applications.??

Non-invasive methods, on the other hand, are less risky and do not require surgery. However, they have limitations in terms of spatial and temporal resolution, which can lead to reduced data quality and information transfer rates. They have significant limitations in their temporal and spatial resolutions such as limitations in Information transfer rate and/or system portability. This makes them less suitable for certain types of BCI applications that require a high level of precision . They have been mostly relegated to tools for basic BCI research and proofs-of-concept, rather than functioning modalities for BCI application.?

Emerging minimally-invasive surgical and high-resolution non-surgical techniques?enrich this landscape.Clinical research involving invasive modalities is extremely important for discovering and gauging the maximum potential of non-invasive approaches.?The utilization of invasive studies to uncover new augmentation and recording techniques for non-invasive modalities leads the way for advancing them. As our knowledge grows, it's becoming easier to identify ideal targets for non-invasive recordings. This helps us to make a shift towards non-invasive BCI modalities that can replace surgical solutions in clinical settings. As researchers focus more on the connections between intrusive, non-invasive, and recovery; enhancement studies, these breakthroughs are bound to become even faster.

As previously mentioned, the development of software to support brain-computer interfacing relies on invasive and minimally-invasive surgical techniques as well as high-resolution non-surgical methods. However, these same techniques are also used by malicious individuals to "brain hack" their victims, causing harm and stealing data. While active BCIs are intended to assist disabled individuals in controlling prosthetic limbs or other devices (see Alimardani et al.2020 ), technological terrorists may use passive BCI methods to monitor victims without their consent, such as detecting cognitive workload, attention, or movements. Passive BCI technology functions in real-time and typically adapts to the user's behavior or response to external events. It is important for the scientific community to develop ethical guidelines and regulations to ensure that brain-computer interfaces are not misused or abused, and to protect against unauthorized access or data breaches.

Terrorist organizations may use invasive neural interfaces on victims as a way to steal confidential information and harm their financial status or reputation. Cyber-attacks by terrorist organizations can be used to manipulate their victims into doing things that are not in their best interest. This could even mean going against their own set objectives as a way for malicious members of the terrorist organization to achieve what they have in mind.Furthermore, the victim will experience continuous intrusion into his private sphere, resulting in the disappearance of his privacy. (Bonaci et al. 2014).

The extraction of private and sensitive information from the human brain has become a chilling reality in today's world (see Ienca et al.2016 ). The groundbreaking research conducted by Martinovic et al. (2012) serves as a testament to this fact. Their research involved presenting EEG brain computer interface users with various stimuli, such as PIN code digits, bank-associated photos, names of the months, debit card digits, locations, and faces. Within each category, a target stimulus, known only to the user, was mixed with non-target stimuli in a random order. For example, in the bank experiment, the target stimulus was a photo of an ATM machine from the user's bank, while the non-target stimuli were images of ATMs from other banks. The aim of the study was to identify a P300 signal in response to private and sensitive information about the user, such as their pin codes, bank membership, month of birth, debit card numbers, home location, and faces of recognized individuals, and extract that data. Hackers could use this information for illegal activities, such as monetary transactions, home banking, or accessing private online accounts. In fact, some terrorist organizations have invested in developing 'brain-spyware' that can extract information directly from brain signaling, enabling them to engage in various criminal activities such as password cracking, identity theft, phishing, and fraud. The potential applications of this technology are vast and can have serious implications for future cybercrime.

Terrorist organizations have found a new weapon in their arsenal: brain hacking. By exploiting the vulnerabilities of their victims, they can gain access to confidential information, manipulate behavior, and steal resources for their own gain. Brain hacking is a particularly insidious form of cybercrime, as it targets the very core of a person's identity and can have devastating consequences. In this chapter, we have explored the motives behind the actions of these organizations and the methods they use to carry out their attacks. In the next chapter, we will delve deeper into the technology behind neural interfaces and how it is used to carry out wide-scale attacks. We will also examine the dangerous tendency to misdiagnose the effects of these attacks as psychiatric disorders, further compounding the harm done to victims. It is clear that brain hacking poses a serious threat to our society, and we must remain vigilant in the face of this new form of terrorism.

?List of references:

Alimardani, M., & Hiraki, K. (2020). Passive Human Brain-Computer Interface for Enhanced Human-Robot Interaction. Frontiers in Robotics and AI, 7, 562611. https://doi.org/10.3389/frobt.2020.562611

Ahmed, Z., Reddy, J. W., Malekoshoaraie, M. H., Hassanzade, V., Kimukin, I., Jain, V., & Chamanzar, M. (2021). Flexible Optoelectronic Neural Interfaces. Current Opinion in Biotechnology, 72, 121–130. https://doi.org/10.1016/j.copbio.2021.04.006

Bonaci, T., Calo, R., & Chizeck, H. J. (2014). App Stores for the Brain: Privacy & Security in Brain-Computer Interfaces. In IEEE International Symposium on Ethics in Science, Technology and Engineering (ETHICS) (pp. 1–4). IEEE. https://doi.org/10.1109/ETHICS.2014.6893177

Hasani, R., Lechner, M., Amini, A., Sichani, M. K., & Maass, W. (2022). Closed-Form Continuous-Time Neural Networks. Nature Machine Intelligence, 4(10), 992–1003. https://doi.org/10.1038/s42256-022-00556-7

Hasumoto, R., Nakadai, K., & Imai, M. (2020). Reactive Chameleon: A Method to Mimic Conversation Partner’s Body Sway for a Robot. International Journal of Social Robotics, 12(1), 239–258. https://doi.org/10.1007/s12369-019-00557-4

Ienca, M., & Haselager, P. (2016). Hacking the brain: brain–computer interfacing technology and the ethics of neurosecurity. Ethics and Information Technology, 18, 117-129.

Lu, W., Zheng, Q., Xu, N., Feng, J., & DTB Consortium. (2022). The Human Digital Twin Brain in the Resting State and in Action. arXiv preprint arXiv:2202.08095.

Martinovic, I., Davies, D., Frank, M., Perito, D., Ros, T., & Song, D. (2012). On the Feasibility of Side-Channel Attacks with Brain-Computer Interfaces. In USENIX Security Symposium (pp. 143–158). USENIX Association. https://www.usenix.org/system/files/conference/usenixsecurity12/sec12-final149.pdf

Polykretis, I., Tang, G., Balachandar, P., & Michmizos, K. P. (2022). A Spiking Neural Network Mimics the Oculomotor System to Control a Biomimetic Robotic Head Without Learning on a Neuromorphic Hardware. IEEE Transactions on Medical Robotics and Bionics, 4(2), 520–529. https://doi.org/10.1109/TMRB.2022.3155278

Wang, Z., She, Q., Smeaton, A. F., Ward, T. E., & Healy, G. (2020). Synthetic-Neuroscore: Using a Neuro-AI Interface for Evaluating Generative Adversarial Networks. Neurocomputing, 405, 26–36. https://doi.org/10.1016/j.neucom.2020.04.069