Organoid intelligence: society must engage in the ethics

The human brain is an amazing organ, far outstripping any current computational devices in processing complex information. Researchers writing in Frontiers in Science have presented a collaborative program to use three-dimensional brain cell cultures known as brain organoids within revolutionary biological computing systems (1). These organoids could also provide a much-needed model system for understanding the brain and brain disorders such as dementia (including Alzheimer’s disease) and schizophrenia. They could even provide a partial understanding of the differences in cognitive consciousness between individuals. 

This ambitious endeavor comes with scientific and technological challenges, but beyond these what are ethical and legal questions for scientists, policymakers, and society at large?

Commentary

An organoid is a miniature three-dimensional structure derived from human cells that are chosen for organ-specific characteristics and which organize themselves to mimic the complex structural and basic functional properties of specific organs or tissues (2). De Jongh et al have extensively discussed the ethical issues associated with research on many types of organoids (2). Like Smirnova et al. (1), I will focus here on brain organoids, which are associated with specific ethical concerns in light of the possibility that they might develop cognitive properties.

Firstly, what benefits do brain organoids promise as an experimental model? During the 20th century humanity reached a consensus that it was not acceptable to experiment on people where significant harm could arise unless the risks and benefits had been properly assessed and subjects gave voluntary informed consent. Indeed, the performance of medical experiments on human subjects without explicit consent was the basis of war crimes trials held in the aftermath of World War II (3). International ethical standards developed in the latter half of that century (4) protect people from the kind of earlier experiments that in many cases were “unconscionable basic violations of ethics even as judged against the researchers’ own recognition of the requirements of the medical ethics of the day” (5).

While it is obviously acceptable to use people who have given explicit consent at late stages of such research, earlier stages of research require experimental models whose characteristics closely resemble those of people. For example, enormous numbers of animals are used instead of human volunteers in the ‘preclinical’ testing of new drugs. However, since animal models may not show the same responses as humans they are often of limited experimental value. The development of organoids from human pluripotent stem cells is likely to constitute a major advance in therapeutic medicine because the organoids’ responses should be a better indicator of peoples’ responses.

Questions of human dignity

Tissue removed from a human being is primarily thought of as garbage and generally disposed of as medical waste. ­­­­The patient, therefore, would not have rights to any research or commercial application through the retention of such garbage (6). This view has changed over the last few years, particularly in the United States with respect to embryonic tissue.  The question of when does an embryo achieve a consciousness as an ‘independent human being’ is hotly debated in many (particularly Christian) cultures, with many people arguing that the potential human being with rights identical to that of its parents occurs at conception.

Arguably the main issue in addressing the ethics of brain organoid use is human dignity – both that of the donor of the cells used to create the organoid, and that of the organoid itself, if it has dignity. If tissue, including cells from which organoids are derived, is retained for research purposes or to develop commercial products does the donor deserve recognition (including an intellectual property component) given that DNA analysis would clearly identify the individual donor? Does the organoid retain the dignity associated with the donor to any degree? If so, how would society choose to protect the dignity and integrity of this clump of cells? Could the organoid experience pain? How would one know, since currently we do not have a baseline for what constitutes consciousness. Can we use organoids in a mechanistic way without ourselves being aware of their state of consciousness?

There are additional questions relating to the donor. For example, what factors could, or should, determine the choice of donor when creating the organoid? Would one choose a donor known to be intellectually strong or weak? What are the implications of the donor selection for organoid characteristics and hence biases in research and any future application of resulting technologies? These considerations relate in part to the issue of bias in artificial intelligence, but perhaps are particularly prescient because of the human origin of brain organoids.

Organoid intelligence vs. consciousness

Whilst organoids mimicking many other human organs would primarily be used for research into the function of the tissue and its response to putative therapies, research into brain organoids may also include their incorporation into some form of biological-machine interface. Smirnova et al. argue that biological computing systems based on brain organoids could be faster, more efficient, and more powerful than silicon-based computing and artificial intelligence (AI) (1). They suggest that such systems could differ significantly in the speed of computation and the energy required to achieve a particular computational result (an important consideration given the vast energy requirements of existing supercomputers) – owing to differences in the ways in which computational decisions are reached.  This would be an exciting use of organized human tissue if and when the myriad connections between the organoid and computers can effectively be achieved. 

Smirnova and colleagues use the term ‘organoid intelligence’ (OI) (1) to encompass this research, likening this term to AI –  the enablement of computers to perform tasks normally requiring human intelligence (7). There have been arguments that computers are already at the point of showing ‘consciousness’ or ‘sentience’; see for example the dismissal of Blake Lemoine, a Google employee, in June 2022 for proposing this (8). But what constitutes these capacities in this context? This brings us back to the key question: are brain organoids conscious, or could they attain consciousness? And if these computer-human constructs were sentient, what rights would they have?

What role for society?

We are entering a new world, where the interface between humans and human constructs blurs distinctions. Society cannot passively await new discoveries; it must be involved in identifying and resolving possible ethical dilemmas and assuring that any experimentation is within ethical boundaries yet to be determined. Society should engage with research on organoid–machine interactions at this early stage for we cannot wait for developments that indicate the personality of the organoid (if any). We must learn from mistakes made in commercializing other new technologies where public acceptance was compromised (e.g., genetically modified crops). This means ensuring that ethical issues acceptable to society at large, and public knowledge and inputs, are embedded throughout these research endeavors. It is critical that a wide range of expertise is taken into account both through embedding and by taking care to involve all stakeholders in the eventual product including the regulatory authorities, industry, those individuals most likely to be impacted by the new technology

Much of this science will result in commercial applications of the organoid-machine hybrids. There are many opportunities to patent and commercialize numerous aspects of the research. There is much to think about as to how this can best be accomplished and whether the donor of the cells has rights. Indeed, if the system is sentient and at least partially human, can ownership rights be enforced?

This short introduction is not intended to provide answers, but rather to highlight some of the many questions related to brain organoid research. We should all be involved in the many decisions which will have to be addressed. 

https://doi.org/10.25453/plabs.22317001.v1 

References

1. Smirnova L, Caffo BS, Gracias DH, Huang Q, Morales Pantoja IE, Tang B, et al. Organoid intelligence (OI): the new frontier in biocomputing and intelligence in-a-dish. Front Sci (2023) 1:1017235. doi: 10.3389/fsci.2023.1017235

2. de Jongh D, Massey EK; VANGUARD consortium; Bunnik EM. Organoids: a systematic review of ethical issues. Stem Cell Res Ther (2022) 13:337. doi: 10.1186/s13287-022-02950-9

3. United States Holocaust Memorial Museum. United States of America v. Karl Brandt et al., Nov. 21, 1946 - Aug. 20, 1947 https://collections.ushmm.org/search/catalog/irn504191

4. World Medical Association. WMA Declaration of Helsinki - ethical principles for medical research involving human subjects (1964/2013). https://www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects/#:~:text=The%20World%20Medical%20Association%20(WMA,identifiable%20human%20material%20and%20data

5. Presidential Commission for the Study of Bioethical Issues. "Ethically Impossible" STD Research in Guatemala from 1946 to 1948 (2011) https://bioethicsarchive.georgetown.edu/pcsbi/node/654.html

6. Court Listener. Moore v. Regents of University of California, 793 P.2d 479 (Cal. 1990) - California Supreme Court (1990) https://www.courtlistener.com/opinion/2608931/moore-v-regents-of-university-of-california/

7. Smirnova L, Caffo BS, Gracias DH, Huang Q, Morales Pantoja IE, Tang B, et al. Organoid intelligence (OI): the new frontier in biocomputing and intelligence in-a-dish. Front Sci (2023) 1:1017235. doi: 10.3389/fsci.2023.1017235

8. BBC news. Blake Lemoine: Google fires engineer who said AI tech has feelings (2022). https://www.bbc.co.uk/news/technology-62275326