Notes on Films from the Future (2018), Chapters 6–7 (batch FFTF-C)#

Source: working/maynard/book/FFTF-C.txt, read in full (PDF pp.109–152). Page numbers below are the PDF page markers, which match the printed pages. Where a quote runs across a page break, both pages are given.

Provenance. The chapters are Andrew Maynard’s own prose, so they count as evidence of his thinking. The plot retellings are his too, but these notes weight his analysis. Some text inside the chapters is quoted from other people and is not evidence of his views: the film epigraphs and dialogue, Joseph Stiglitz (p.110), the Pew survey (p.123), Prellis’s stated goal (p.117), Elon Musk’s tweet and Neuralink’s recruitment line (p.128), Emily Yoshida (pp.129–130), Aimee Mullins’s TED talk (p.137), the IAAF rule (pp.137–138), Hugo Campos (p.147), England et al. (pp.148–149) and the Muddy Waters report (pp.149–150). He chose and framed these quotes, and I note how he uses them. Quotes below keep the source’s spelling, including its typos (marked [sic]).

Chapter 6: Elysium: Social Inequity in an Age of Technological Extremes (pp.109–127)#

The argument#

Elysium is a film Maynard calls “rather earnest” (p.109) and “a rather naïve movie” (p.114). He uses it to argue that powerful technologies do not spread their benefits evenly on their own. Unless they are “developed and used responsibly”, they widen the gap between the privileged and everyone else (p.110). The chapter moves through three kinds of technology-linked inequity:

1. Unequal access to benefits (pp.110–118). He reads the film’s medical pods as a metaphor. Real bioprinting (Atala’s 2016 ear, Prellis’s holographic printing) shows the promise is partly real, which raises the question of who will get it. The risk he names is a “two-tiered society” in which the rich upgrade their bodies and the poor watch (p.118).
2. The disposable workforce (pp.118–123). This is the chapter’s analytical core and the part most rooted in his own career. Advanced technology does not make workplaces safe. Scientific uncertainty, profit pressure and plain greed let employers push risk onto workers. Harm is displaced rather than removed: offshored, or loaded onto the “first tier” of people who handle new materials.
3. Automation and education (pp.123–127). In the film, automation “didn’t deprive them of a job, but it did deprive them of choice” (p.123). In the real world, whether automation cuts jobs overall is still an open question. What is fairly sure is that jobs will be displaced, so the deciding factor is whether people can reskill. That makes education, rethought and made far more widely available (YouTube, Crash Course, ASU’s online model), part of “the bedrock on which social equity is built” (p.127).

His conclusion is that education is necessary but not enough. Avoiding an Elysium future “will also require a deep commitment to developing, using, and governing new technologies responsibly and ethically” (p.127).

Concepts and frameworks#

Risk#

• Occupational and first-tier risk is the lens here, and it comes straight from his professional life. He spent 13 years in workplace safety at the UK Health and Safety Executive (from 1992) and at NIOSH, doing bench research on airborne dust (p.118). His grandfather died of coal miner’s pneumoconiosis in 1977 (p.119). On one of his last visits down an English pit he was there to check that dust samples were not being tampered with (p.120).
• Uncertainty as a political resource. Scientific uncertainty over black lung delayed action, and it “suited the mine owners” (p.120). He also cites the 1991 US dust-sample tampering scandal, in which more than 800 mines were implicated (p.120). The lesson he draws is that it was “all too easy to cut corners in order to increase production” (p.120), even though the protective technology existed. Black lung is “once again on the rise” (p.121).
• Tradeoffs between safety and productivity are built into the system. “for many people, a dangerous job is better than no job at all” (p.121). He cites US and global burden figures (Takala et al.: 2.3 million work-related deaths in 2012) as “certainly not what might be hoped for in a technologically advanced society” (p.121).
• Emerging technologies carry the same pattern. Nanomaterials and GM microbes in bioreactors show “relatively little correlation between the sophistication of the technology and the safety of the environment in which it’s used” (p.121). In footnote 73 (p.122) he notes that despite about two decades of research, “some companies continue to use these as if they are, by default safe”, and that carbon nanotubes “may, under some circumstances, behave like asbestos if inhaled”. He cites his own 2016 Conversation piece, “We don’t talk much about nanotechnology risks anymore, but that doesn’t mean they’re gone”.
• How likely is a disposable workforce? “If we don’t take workplace health and safety seriously, and the broader issues of social justice that it’s a part of, I’m sad to say that it’s pretty likely” (p.122). He frames workplace risk as part of social justice, not only a technical matter.
• Public risk perception. He treats the Pew findings “with a pinch of salt”, since they are “opinions rather than predictions” (p.123). He points to the paradox that 56% would not ride in a driverless car for safety reasons, although safety is the main argument for automation (p.123). Footnote 76 cites his 2016 piece asking whether driving your own car will become as socially unacceptable a public health risk as smoking. He grants that public worry about automation and jobs rests on “impressions and gut feeling rather than hard evidence”, yet is “probably justified in one respect” (p.124). He takes lay intuition seriously without deferring to it.
• Risk communication (autobiographical). Footnote 80 (p.126): “In 2012, I launched the YouTube channel Risk Bites as a platform for helping people make sense of risk, including the potential risks and benefits of emerging and converging technologies.”

Responsible and socially responsible innovation#

• Responsibility is the chapter’s frame from the first page: technology can widen divides “if not developed and used responsibly” (p.110), and “powerful technologies come with great social responsibility” (p.111).
• He names “corporate social responsibility and responsible innovation” as emerging ideas that “help here”, along with “innovative corporate leadership that actively strives to reduce social inequity” (p.122). Footnote 74 gives B Corp certification as an example of “socially responsible corporate leadership” (p.122).
• The business case: “The economics of accident and disease prevention make good business sense, as do the economics of fostering a happy and thriving workforce” (p.122). Firms that go “beyond regulatory requirements” are good news (p.122).
• Its limits: “the fiscal temptation to use cheap labor is sometimes a tough one to resist” (p.122). So responsibility cannot sit only with firms or regulators; it is “the responsibility of everyone” (p.122).
• Closing formula: “a deep commitment to developing, using, and governing new technologies responsibly and ethically” (p.127). Responsible innovation is invoked but not defined in this chapter. It works as one lever among several (CSR, leadership, regulation, education) aimed at equity and social justice.

Permissionless innovation#

The term does not appear. The closest theme is his account of industry treating new materials as “by default safe” (fn73, p.122) and taking risks with workers’ health for profit.

Hubris#

The word does not appear in this chapter. A related point: the film’s single-author operating system, which its creator can still hack, is a “jaw-dropping level of vulnerability” (p.112). Maynard notes it only in passing, as plot, but it prefigures Chapter 7’s point about systems built without attention to what could go wrong.

The people who develop technologies and their mindsets#

• Employers and CEOs: Carlyle (the film’s CEO) is the type of the profit-first owner. In the real world, “mining corporations and manufacturers couldn’t afford to acknowledge they might have a problem” (p.120).
• Scientists: he writes admiringly of Atala’s team and of Prellis. His own stance is skeptical realism about hype: “I’m a little skeptical about how soon we’ll be able to print replacement body parts on demand, as biology is constantly blindsiding us with just how deeply complex it is” (p.117). This sits alongside real enthusiasm: “This is quite amazing stuff” (p.117).
• Workers and cheap labor: some people “are willing to work for less and cut corners to get ahead of their peers” (p.122), which is part of why responsibility is shared.
• Academics and experts: he criticizes his own profession for being absent from online learning (p.127). Footnote 79 (p.125) endorses ASU and Michael Crow’s model of a public university that keeps up with the technological future: “It’s why I work here.”

Governance and public engagement#

• Regulation appears as necessary but open to gaming: mines struggled “to comply with exposure regulations” while staying solvent (p.120), and samples were tampered with (p.120). Recognition of the problem and regulation of UK mines came late, in the 1940s (p.120).
• Governance is widened beyond regulators to “everyone” (p.122) and to the joint work of “developing, using, and governing” (p.127).
• Public engagement shows up mainly as attention to public attitudes (Pew) and as a call to democratize knowledge: “knowledge will no longer be the domain of those rich enough to afford it, or privileged enough to use it” (p.127). Access to knowledge is treated as a condition for social equity.

AI and intelligence#

This is a minor strand in Chapter 6, and it appears mostly through the film. - AI and robots are shown as instruments of stratification: “subservient AI robots” serve the rich (p.110), while “zero-tolerance autonomous-robot law enforcement” (p.111) and a robot parole officer (p.112) police the poor. The irony he stresses is that the disposable workers are building “robots that are designed to further suppress the earth-bound population” (p.122). - Automated gatekeeping of rights. Elysium’s operating system “defines and oversees all of the orbital’s operational and social functions, including who is a citizen” and “who has the authority to govern” (p.112). The medical pod refuses a non-citizen child (p.114). This is plot, but it frames code as a constitution that sets who gets access and power. - Pew respondents “didn’t like the idea of computers deciding who to hire” (p.123). - On automation and jobs he relies on Brynjolfsson and McAfee’s open question of net job loss versus changing jobs (p.124), and uses the Luddites as the historical case (p.124).

Technology convergence#

The theme is implicit. Bioprinting is 3-D printing combined with cell biology (pp.114–117). Neural interfaces are dismissed as a “technology of convenience” in the film (p.112). Risk Bites is described as covering “emerging and converging technologies” (fn80, p.126).

Analogies across technologies#

Key quotes (Chapter 6)#

• “science and technology can, if not developed and used responsibly, deepen the divide between those who live comfortable, privileged lives, and those who do not” (p.110)
• “powerful technologies come with great social responsibility” (p.111)
• “not amenable to simple, quick fixes” (p.114, on the roots of social inequity)
• “just because we can cure the sick, that doesn’t mean that everyone will benefit” (p.117)
• “manufacturing at scale is often dirty and dangerous, even if the technology being manufactured is not” (p.118)
• “This was an uncertainty that suited the mine owners” (p.120)
• “there is relatively little correlation between the sophistication of the technology and the safety of the environment in which it’s used” (p.121)
• “one of the greatest threats to workplace safety is not a lack of technological innovation, but ignorance of what might go wrong with novel technologies” (p.122)
• “preventing a future disposable workforce becomes the responsibility of everyone, not just employers or regulators” (p.122)
• “automation didn’t deprive them of a job, but it did deprive them of choice” (p.123)
• “we could be on the cusp of an unprecedented democratization of education” (p.127)
• “meaningful access to knowledge and understanding for all is part of the bedrock on which social equity is built” (p.127)

Chapter 7: Ghost in the Shell: Being Human in an Augmented Future (pp.128–152)#

The argument#

Ghost in the Shell (1995) is a film Maynard clearly loves (“this meditation that’s masquerading as an anime movie”, p.131). He uses it to ask what being “human” means once machine augmentation blurs the line between biology, machines and cyberspace (pp.128–129). The chapter has four linked moves:

1. Identity and the “ghost” (pp.129–134). Neural implants make identity hackable: hackers “implant false memories in people’s minds, altering who they think they are” (p.129). Kusanagi’s doubt about whether she is who she thinks she is sets up the question of what gives an augmented being worth. The film’s answers are diversity (“overspecialization leads to death”, p.131), relationships built on “respect, acceptance, and empowerment” (p.132), and letting go of the past (p.133).
2. Social risks of augmentation: “normal” and “human” (pp.134–141). Steve Mann’s EyeTap incident, body hackers, transhumanists, and the contrast between Pistorius and Mullins show that societies welcome enhancement only until it threatens their sense of how the world should be. They then slide from “abnormal” to “not human”. His prescription is to move from socially lazy, instinctive assimilation to intentional, value-guided frameworks.
3. Neural augmentation and hacking (pp.141–146). We tie identity to the brain, so threats to it are felt as major risks, and neural interfaces bring “a brand new set of threats to self-identity”, including “vulnerability to outside manipulation” (p.142). Cyberspace is a human-made layer built without much thought for what could go wrong. As the cyber and physical worlds converge, you can no longer just pull the plug, and a wired brain opens “a one-way neurological door” into a world “that reflects all the limitations and biases and illusions that come with human hubris” (p.146). He does not say don’t build. He says build with maps marked “Here be monsters” (p.146).
4. “Who owns you?” (pp.146–152). Kusanagi’s company-owned body, Campos’s locked-out defibrillator data, the England et al. deactivation case, the St. Jude and Abbott firmware recall, and the 2016 IoT botnet attacks all show that intimate implants tie users to manufacturers for access, upgrades and security. For enhancements, as opposed to medical devices, we face “an ethical and moral vacuum” (p.149). Regulation will have to adapt, and consumer demand for upgrades may push back against it (p.152).

The chapter hands off to Ex Machina by noting that the alternative, “creating machines that can achieve what we only dream of”, brings “its own challenges” (p.152).

Concepts and frameworks#

Risk#

• Social risk of augmentation. “these emerging technological capabilities come with a complex array of risks, as Steve Mann’s experience showed” (p.136). The risk he means is social strife and discrimination, not physical harm: “If we want to see the rise of increasingly augmented individuals, we need to be prepared for some social strife” (p.136).
• Perceived and actual risk. He calls for “a society-wide appreciation of the perceived and actual risks and benefits of augmentation and enhancement” (p.140). He keeps both perception and “actual” risk in view.
• Risk as threat to something valued. Threats to lifestyle, possessions, beliefs and identity trigger “a metaphorical circling of the wagons” (p.139). Identity is so precious that threats to it are “perceived as a major risk” (p.142). The chapter does not state a formal definition of risk, but the reasoning treats risk as a threat to what people value.
• Unknown risk in frontier technology. Neural lace is “a technology with both vast potential and largely unknown risks” (p.142). “we’re still far from understanding the basics of how it operates… Yet this is not stopping people from experimenting, despite what this might lead to” (p.142).
• Hacking as near-certain. “there is near-certainty that, if and when someone connects a part of their brain to the net, someone else will work out how to hack that connection. This is a risk that far transcends the biological harms that brain implants and neural nets could cause” (pp.145–146). He then adds “perhaps an even greater risk”: entering an unevolved, hubris-laden designed world (p.146).
• Systemic interdependence. Cyber and physical convergence means “cyber-vulnerabilities have sometimes startling real-world consequences”, from thermostats to pacemakers to brain-machine interfaces (p.145).
• Legacy and lock-in risk. The 2016 botnet attacks on “dumb devices” show that the only sure fix for vulnerable hardware may be to “physically replace millions—probably billions—of chips” (p.151). For implants, this raises the question of who is “responsible for its security” (p.151).
• Safety of DIY enhancement. “The ethics of untrained and unsupervised people cutting themselves and others open to insert objects of unknown provenance are interesting to say the least, never mind the safety concerns” (pp.134–135).

Responsible and socially responsible innovation#

• He adds “ethical and socially responsive development and use” to Campos’s point: without it, “a user’s identity becomes connected to those that have control over the device and its operations” (p.147).
• The positive program is intentional and values-based: “working out what’s important, and intentionally opening pathways for this to occur” (p.140); frameworks guided by “socially-agreed-on values” (p.141); “plenty of soul-searching around our collective values, and how we put them into practice” (p.140); rethinking “personal value, worth, and rights” (p.141).
• The goal is framed in positive terms: “the beneficial, equitable, and—let’s be honest, life-enhancing and affirming—development of augmentation technologies” (p.140).
• The words “responsible innovation” do not appear in this chapter. The ideas behind them (anticipation, values, equity, duty of care) do.

Permissionless innovation#

The term does not appear, but the chapter is full of cases where innovation outruns oversight: - Body hacking as “this extension of the maker movement”, often “through unsupervised do-it-yourself surgery” (p.134). - Phil Kennedy’s elective brain-implant surgery on himself (p.143). - Early computing: “we were more interested in what we could do rather than what we should (or even how we should do it)” (pp.144–145). - “poorly-regulated innovation around body-implantable enhancements that aren’t designed for medical reasons, but to enhance ability” (p.151). - A demand-driven push against oversight: “in the rush for short-term gratification, we’ll see pushback against regulations that make it harder for consumers to get the upgrades they crave” (p.152).

His stance is not prohibitionist: “Not that I think this should be taken as an excuse not to build brain-machine interfaces. And in reality, it would be hard to resist the technological impetus pushing us in this direction” (p.146). He sees technological momentum as a fact and argues for navigation aids (maps, frameworks, adaptive regulation), not stopping.

Hubris#

• The one explicit use, and the key one: the cyber world “reflects all the limitations and biases and illusions that come with human hubris. This is a world wherein human-made monsters lie waiting for us to stumble on them” (p.146). The idea is structural. Engineered systems lack the long error-correction of evolution, so designers’ blind spots become built-in hazards.
• Implicit hubris shows up in several places: the belief that “someone could always just pull the plug” (p.145); “making up the rules along the way, and not worrying too much about what could go wrong until it was too late” (p.145); the creators of algorithm 2501 who “thought they had it under control” (p.132, plot retelling); and transhumanist framings that equate merging with machines with evolution, which he calls “a stretch” (p.135).

The people who develop technologies and their mindsets#

• Tech entrepreneurs. Musk’s neural-lace tweet “might just have been a bit of entrepreneurial frippery” (p.128). On Neuralink’s line “No neuroscience experience is required”, he says “it’s a little scary to think that a bunch of computer engineers and information technology specialists could be developing advanced systems to augment the human brain” (p.128). More broadly, “entrepreneurs and technologists become ever more focused on fixing what they see as the limitations of our biological selves” (pp.128–129). The worry is engineering confidence working outside its field, combined with a deficit model of biology.
• Body hackers and transhumanists. Body hackers are “often enamored with the idea that we can use technology to overcome our biological limitations, and transcend our evolutionary heritage” (p.135). He names Kurzweil, Bostrom and Zoltan Istvan (p.135) and describes rather than attacks them, but pushes back on the evolutionary framing (p.135).
• Scientist self-experimenters. He calls Kennedy’s novel 2051 “rather fanciful” but says it “does hit on” the hacking problem (p.144).
• Early computing culture. Could over should (pp.144–145). The digital community did learn that cybersecurity “demanded at least as much attention to good practices, robust protocols, smart design, and effective governance as any physical environment” (p.145).
• Manufacturers. They hold data, upgrades and security, so they hold power over users (pp.147–151). The Muddy Waters report and FDA recall are presented without taking sides on St. Jude’s denial (p.150).
• Industry insiders on risk. A senior IBM executive told him “cybersecurity is one of the greatest challenges we face as a global society” (p.144).

Governance and public engagement#

• Society-level deliberation about values (pp.140–141), not only expert regulation.
• Regulation: “tough medical device regulations have limited the extent of potential problems here so far” (p.151). Regulation for enhancement must “change and adapt to keep up”, ideally with laws protecting “consumers’ quality of life for the duration of having machine enhancements surgically attached or embedded” (p.152).
• Rights and autonomy: who can switch off or change an implant, whether you can refuse an overnight upgrade, your legal rights when a buggy patch causes “a quite-literal brain freeze”, and what happens with unsupported legacy implants (p.149).
• Anticipatory attention: “will anyone pay attention before we’re plunged into an existential crisis around who we are, and who owns us?” (p.152). He holds that film can offer “prescient foreshadowing” of legal and social challenges (p.152).
• Sport as an example of governance under threat: the IAAF’s reaction to Pistorius shows institutions guarding “normal” (“You could almost hear the IAAF breathe a collective sigh of relief when he didn’t win”, p.138).

AI and intelligence#

• Algorithmic manipulation and loss of control. In the film, the Puppet Master is “an algorithm—codenamed project 2501—designed to hack people and cyber-systems and manipulate them” that “became self-aware and escaped out into the net” (p.132, plot retelling). Its request for asylum “as a life-form” (p.133) raises, without resolving, the question of the moral status of an artificial mind.
• Benevolent-AI manipulation. In Kennedy’s 2051, AI nanobots can “manipulate someone’s reality with ease, and even create an alternate reality that they are incapable of perceiving as not being real”. They want “global peace and universal happiness” and try to assimilate humans, which he calls “Borg-like… but with a benevolent twist” (p.144). He calls the plot “rather silly” but takes seriously its core point about vulnerability to manipulation.
• Ceding power to autonomous systems. Hacking is most challenging “in a world where we are seceding [sic] increasing power to autonomous systems” (p.144).
• Cognitive augmentation. Neural lace could “turbocharge your cognitive abilities”, “change your mood, recall information faster, get real-time feedback on who you’re speaking with” (p.142). It is seductive, and he is doubtful it will happen soon (pp.136, 142). This links back to Limitless (ch.5) on cognitive enhancement (p.139).
• Manipulation of mind and self. This is the part of the chapter most relevant to his later concerns about manipulation and cognition. Identity can be rewritten (“They can, in effect, rewrite who someone thinks they are”, p.130). He counts “persuasive influences” alongside disease and injury as forces that can change who we are (p.142). The fear is “becoming someone else’s puppet” (p.145), and Kusanagi’s worry is that she might not even know if she had been compromised (p.145).

Technology convergence#

This is explicit and central to the chapter: - “As technological capabilities have continued to grow and converge in areas as diverse as robotics, nanotechnology, AI, neurotechnology, and biotechnology…” (p.135). - “Advances in AI-related technologies, computing architectures, gene editing and manipulation, robotics, on-demand additive manufacturing, and the converging and merging of these and other technologies, is massively accelerating what is possible” (pp.135–136). - “It’s this convergence between cyber and physical realities that is massively accelerating current technological progress” (p.145). Convergence both drives capability and spreads vulnerability. - “the boundaries between biology, machines, and cyberspace are becoming increasingly blurred” (p.129).

Analogies across technologies#

Key quotes (Chapter 7)#

• “it’s a little scary to think that a bunch of computer engineers and information technology specialists could be developing advanced systems to augment the human brain” (p.128)
• “This possibility for mind and memory manipulation gets to the heart of Ghost” (p.129)
• “They can, in effect, rewrite who someone thinks they are” (p.130)
• “As a species, we are embarrassingly programmed to see “different” as “threatening,” and to take instinctive action against it” (p.136)
• “We’re also going to have to grapple, perhaps more than in any previous technological age, with what it means to be “human” as we artificially augment ourselves” (p.136)
• “As a species, we’re remarkably good at celebrating success, as long as it doesn’t undermine our sense of how the world should be” (p.138)
• “There’s a dark, deep tendency to label threateningly different traits and abilities as “non-human” or even “sub-human” in our collective psyche” (p.139)
• “But it’s also socially lazy. It’s an assimilation that lacks consideration and intentionality” (p.140)
• “neural enhancements bring with them a brand new set of threats to self-identity” (p.142)
• “a technology with both vast potential and largely unknown risks” (p.142)
• “this pulling-the-plug concept seems like a quaint and hopelessly outmoded idea” (p.145)
• “in our eagerness to connect our very soul to the net, we risk losing ourselves, or worse, becoming someone else’s puppet” (p.145)
• “This is a world wherein human-made monsters lie waiting for us to stumble on them” (p.146)
• “Not that I think this should be taken as an excuse not to build brain-machine interfaces” (p.146)
• “The company may not own the hardware, but without a doubt, they own you, or at least your health and security” (p.151)
• “will anyone pay attention before we’re plunged into an existential crisis around who we are, and who owns us?” (p.152)

Cross-chapter threads#

• Power and dependency. Elysium’s operating system decides citizenship. Megatech owns Kusanagi’s body. ICD makers control patients’ data. In each case, whoever controls a technology controls people’s access, rights and even identity.
• Equity. Chapter 6’s two-tiered society returns in Chapter 7 as the question of who can afford upgrades and who is stuck with unsupported, vulnerable hardware (pp.149–151), and as the call for “equitable” augmentation (p.140).
• Realism about hype. He is skeptical of bioprinting timelines (p.117) and of neural lace (pp.136, 142), yet insists “we’re not as far as we sometimes think” (p.136), and he warns against treating technology’s lack of plausibility as a reason to stop paying attention.
• Film as a lens. He admits the films are implausible (the pods; Carlyle’s brain-stored patch as “a technology of convenience”, p.112), yet they surface “social challenges that are deeply relevant to technological innovation in today’s world” (p.114) and offer “deeply prescient foreshadowing” (p.152).
• The lived expertise behind the risk thinking. Chapter 6 is one of the book’s most autobiographical risk chapters: HSE, NIOSH, dust science, his grandfather’s black lung, auditing mines, Risk Bites, ASU.

Digest: what these chapters add to the map of Maynard’s thinking#

1. Risk as a social-justice problem, grounded in occupational health (central and lasting). Chapter 6 shows where much of his risk thinking comes from: thirteen years of workplace dust research at HSE and NIOSH, a grandfather killed by black lung, and a site visit to make sure dust samples had not been tampered with. The lessons he takes from this experience recur across his work. Technical sophistication does not produce safety. Harm lands first on “the first tier of people” who make and handle new technologies (p.121). Scientific uncertainty can be a convenience for those who profit (“an uncertainty that suited the mine owners”, p.120). With novel technologies, the main danger is “ignorance of what might go wrong” (p.122). The link from coal dust to nanomaterials to GM-microbe bioreactors, including the carbon-nanotube and asbestos warning in footnote 73, is his characteristic move of carrying lessons from older hazards to emerging technologies. It is mostly structural, with one literal toxicological strand. This is core to the map. It is the empirical base under his later, more abstract talk of risk, and it frames workplace safety as part of “the broader issues of social justice” (p.122).

2. Responsible innovation as a shared, equity-focused commitment (central, but light on theory here). Responsibility frames both chapters. There is “great social responsibility” (p.111), a need for “developing, using, and governing new technologies responsibly and ethically” (p.127), and “ethical and socially responsive development and use” (p.147). He treats responsible innovation and CSR as useful but limited, sitting alongside business sense, corporate leadership, regulation and education. His distinctive claim is that responsibility is distributed: it belongs to “everyone, not just employers or regulators” (p.122). He does not develop a formal framework here. The weight is on intention and values over instinct.

3. Being human under augmentation, and the social risk of “normal” (central to the book and to his later work on being human). Chapter 7 gives one of his clearest accounts of how societies react to enhancement. People welcome difference until it threatens what they value, and then slide from “abnormal” to “not human” (p.139). Instinctive assimilation is “socially lazy” (p.140), and the remedy is deliberate, society-wide conversation about values, rights and worth. Two ideas here seem lasting. One is that “we are… already a technologically augmented and enhanced species” (p.140). The other is that risk, as people experience it, is a threat to what they value, including identity itself (pp.139, 142).

4. Manipulation and the malleable self (an important seed). Ghost-hacking gives him a way to say that identity is malleable, and that “persuasive influences” belong with disease and injury as forces that can change us (p.142). His fear of “becoming someone else’s puppet” without knowing it (p.145), and the benevolent AI in 2051 that manipulates reality (p.144), point ahead to his later concerns about technologies that shape cognition and selfhood. In 2018 the channel is neural hardware, not language. The underlying worry, outside influence on who we think we are, is the same.

5. Hubris, designed worlds and the pull-the-plug fallacy (moderately central, and pertinent to AI). Cyberspace was built by people asking what they could do, not what they should. It is a human-made “fifth dimension” with no evolutionary debugging, full of “human-made monsters” (p.146). Once cyber and physical systems converge, the “pulling-the-plug concept” is “hopelessly outmoded” (p.145). His response is characteristic: do not stop building, but navigate with maps marked “Here be monsters” (p.146). This mix of accepting technological momentum while insisting on foresight is a stable feature of his stance, and it anticipates later arguments about AI control.

6. Who owns you: dependency, control and regulatory lag (moderately central). The corporate-body analysis (implant data, deactivation rights, forced upgrades, legacy hardware, a continuing duty of care) shows him working through governance at the level of concrete institutions and rights. He expects regulation to fall behind consumer demand (“the rush for short-term gratification”, p.152). The idea returns in his broader writing about power and dependence on technology companies, but here it is tied to implants.

7. Automation, work and education (lasting, secondary in this book). He treats automation as taking away choice more than jobs (p.123), and education as the main lever against automation-driven inequality. That includes criticizing academics for staying out of online learning (p.127) and linking the point to ASU and Risk Bites. His lasting interest in rethinking education and public learning starts here.

Less central here: AI as such (mostly through the films’ robots and algorithm 2501), and permissionless innovation, which is shown through body hacking and DIY self-experiment but never named. Convergence is explicit and treated as the engine of both capability and vulnerability, but as background, not as a developed framework.

Overall. These chapters show a scholar whose risk thinking starts from harm to people, especially the least powerful. He is enthusiastic about technology but skeptical of hype and of the mindsets of engineers and entrepreneurs who work outside their fields. He consistently points toward intentional, value-guided and widely shared responsibility as the alternative to both instinctive reaction and unchecked momentum.