A well-studied source of existential risk is one where people have enough knowledge of the causes and processes involved to predict the occurrence of the existential risk given the presence of a set of causes. By "have enough knowledge of the causes and processes involved", I mean scientific models and limited real-world examples studied over decades of time and with strong consensus about their implications and applicability. By studying a source of existential risk well, you can develop models of how to prevent that source of risk from occurring or develop ideas for how to reverse its consequences.
A predictive indicator of a source of existential risk can be measured and provides warning months or years in advance of precipitating events that are sufficient to cause large percentage losses of human life. Predictive indicators of well-studied sources of existential risk (for example, climate change and nuclear proliferation) are subject to global efforts to detect them and prevent consequences of their source.
Well-studied sources of existential risk with predictive indicators include:
* A set of self-amplifying consequences of climate change.
* An animal epidemic that leaps to humans because of ecosystem encroachment.
* A nuclear exchange that is entirely due to human decisions.
Not well-studied existential risks with predictive indicators include:
* A loss of insect species that has unknown extent and rate.
* A widespread automation failure caused by a solar flare.
* A loss of plankton species that has an unknown extent and rate.
Not well-studied existential risks without predictive indicators include:
* A human organization develops a superintelligence that escapes.
* A human organization develops a gray goo nanoweapon that is released.
NOTE: leading indicators of work on advanced AI or nanoweapons are available to the world's intelligence agencies, assuming that they know how to interpret their evidence.
Sources of existential risk that are well-studied and have known and monitored predictive indicators have a special status if those sources of existential risk are growing progressively more harmful despite decades of study and monitoring. Climate change is an example of such a source.
Sources of existential risk that are not well-studied but could be growing progressively more harmful and do in principle have predictive indicators include critical species losses. The decline of insect species on land will kill most humans. That decline can be measured and protected against, even reversed, in advance of its consequences.
Understand when your quality of life is threatened by a risk that is common for all people on the planet. The sources of existential risk that I listed, and others that I didn't, involve a threat to quality of life for every human on the planet and some sources of risk (for example, nanotech grey goo) threaten the existence of all life on the planet. By the end of the century, all the sources of risk that I listed can finish their consequences for human life (yes, the gray goo as well).
To summarize, some sources of existential risk:
* might have predictive indicators
* can be studied
* can be prevented
* might require a human choice to make them occur
* can finish their consequences this century
Study the sources of risk and find predictive indicators of them to monitor. Then make a plan to prevent the risk. In the case of climate change, even with a well-studied source of risk that has monitored predictive indicators, and even with feasible and frequently updated predictions and plans from the UN IPCC, those plans have not been put into action yet. However, that is not for lack of information about how to proceed.
There is still hope for several more years that a radical change will force compliance with the plans proposed by the IPCC. Beyond that, hopes turn to radical technology improvements and investments that allow control over the causes and consequences of climate change.
A growing concern about climate change is lock-in, that is, when do we lock-in consequences for the planet that just take time to manifest. We can cause a result through gas release to the atmosphere that has consequences that are only observable decades later. Lock-in is a concern with species extinction as well.
The climate change community now proposes models of human social or societal tipping points, positive tipping points. The community suggests that positive tipping points model changes in human response to climate change over time. This is what to hope for now.
Response to negative short-term events could trigger positive change as well (for example, a global response to a tropical heat wave reaching lethal wet bulb temperatures for a few weeks). The trouble with relying on an event like a widely lethal heat wave to "wake people up" is that it will involve massive loss of life and resources. Furthermore, the event could be just enough to destabilize global society (for example, by triggering a global migration crisis, a few wars, a global famine, and a weakening of global power structures that causes lack of global government coordination to prevent further consequences of that one heat wave). Our poor collective response to predictable negative consequences with weak direct existential risk is a strong indirect existential risk.