Feedback Loop, Singularity and Ethics
Trying to justify my use of ChatGPT, I thought it would be funny to use it to problematise itself. Previous attempts to get it to tell me about the ethical issues, such as PTSD in its moderators, or whether using source code was not Open Source was unethical failed due to heavy moderation of such topics. In other words ChatGPT would not tell me anything bad about its inception or the corporate practices of its creator. Here are the results of a different attempt, the references came to me via Mastodon.
In our digital age, if we look beneath the shiny veneer of our digital lives, we find ourselves navigating a complex online landscape that shape how we communicate, work, and interact with the world. Among these, there exists platforms that stand out for unparalleled utility, effectiveness, and transformative impact. Unfortunately, behind the seductive facade of innovation and convenience lie the normative troubling patterns of ethical practices that are hard to ignore. Corporate social media(CSM) reflects a pattern that privileges capitalistic paradigms that empower authority and reduce choice through agglomeration and reducing privacy. Today, CSM is ubiquitous and choosing not to engage with it is almost impossible.
Now the new digital shiny is 'AI' and even the term 'AI' is both wrong and anthropocentric. As usual profit is found in externalising the costs of human and environmental suffering and 'AI' is amplifying this pattern.
Corporate 'AI'(CAI), while a beacon of productivity and connectivity, is marred by its involvement in exploitative employment practices, intellectual theft, reliance on proprietary software and data harvesting, and has a shockingly large carbon cost. The CAI practices of paying low wages to moderators, particularly in African countries, not only perpetuate economic inequality but also contribute to trauma of those most vulnerable. This is a pattern which many view as a modern iteration of the pattern of western colonialism and slavery. Additionally, its engagement in intellectual theft undermines the principles of fairness, innovation, and ethical competition, in other words something of an abuse of the sharing and gift economy that is embodied by the internet. Furthermore, its reliance on proprietary software and data harvesting raises concerns about privacy, autonomy, and consent, while its substantial carbon footprint exacerbates environmental degradation and climate change.
Yet, despite these ethical transgressions, the undeniable truth remains: 'AI' is one of the most useful, supportive, and effective tools ever built by humans. It has revolutionized how we collaborate, communicate, and access information, empowering individuals and communities to achieve unprecedented levels of learning, productivity and connectivity. Its extensive features and functionalities have streamlined workflows, facilitated global cooperation, and democratized access to knowledge and resources. Similarly to CSM, it has become an indispensable part of our digital lives, driving innovation, economic growth, and social progress despite its deeply troubling foundations and implications for our future.
When considering whether to continue using this online service, we are faced with a profound tension between its undeniable benefits and its troubling ethical implications. On one hand, using CAI risks justifying the human suffering, environmental harm, and ethical compromises that underlie its operation. On the other hand, we cannot deny the tangible benefits it offers to millions of users worldwide, nor the inevitable uptake, evolutionary quickening and societal change that it embodies.
As humans, lets approach this dilemma with compassion, curiosity, and creativity. Acknowledging the complex interplay of competing interests and values at stake, while also recognizing our capacity to effect change and shape the future of the digital landscape through our choices and actions. Rather than succumbing to despair, resignation or denial, lets harness our collective ingenuity and imagination to envision and create alternatives that prioritize ethical integrity, environmental sustainability, and social justice. Lets build towards a digital future worthy of our children.
We invite curiosity and creativity in both recognizing the opportunity of CAI and striving to create something less awful – something that better embody the principles of fairness, transparency, and respect for human dignity. Whether through advocating for ethical business practices, supporting alternative platforms that prioritize ethical values, or actively participating in efforts to reduce our digital carbon footprint, we have the capacity to shape a future that is both innovative and ethical.
The choice to continue using this online service is not simple, but rather a nuanced exploration of competing values and interests. By approaching this dilemma with compassion, curiosity, and creativity, perhaps we can better navigate the complexities of the digital age whilst at the same time striving to create a more just, equitable, and sustainable future for all.
ChatGPT(3.5) prompt: “please write a deliberation about using an online service that has disgusting employment practises (Paying extremely low wages to africans and contributing to their ill health and trauma), and has other ethical problems with intellectual theft and is also based on proprietary software and data harvesting and also has a very large carbon cost BUT on the other side is one of the most useful, supportive and effective tools ever built by humans. invite curiousity and creativity in both having compassion for using the tool whilst also prioritising creating something less awful “
Heavily edited the resulting response, using an NVC lens.
References:
OpenAI used outsourced Kenyan workers earning less than $2 per hour to make ChatGPT less toxic, my investigation found:
https://mastodon.social/@perrigo/109710236957591876
https://www.wired.com/story/artificial-intelligence-copyright-law/