Here's a general thumb rule:
When a business claims something to be science and if that so-called science hasn't been peer reviewed in the scientific community, chances are ... it is pseudo-science. Time to put on your skepticism cap!
1. Understanding growth
What does the world population look like?
That linear ramp from 5000 yrs ago to today... Agriculture started 5000 years ago. Interesting.
The estimated population of the world is around 6,804,600,000. The estimated growth rate of today is a meagre 1%. Well, 1% of 6.8bn is 68 million (68,046,000). That's quite a lot of new people every year. In 5 years... there'll 7 times the number of people there were in 1900.
Will the solution then be to further keep tinkering with our crops' DNA to keep producing ever increasing amounts of food? The second law of thermodynamics doesn't like this... because our only source of energy is the sun. Even if the plants one day become 99% efficient in using the sun's energy, there is no way to ramp up the sun's light.
So... Growth is unsustainable on a finite planet. The argument "We need to feed ever increasing populations" is therefore only an excuse to buy out land and hold onto wealth. Your wealth.
2. Understanding the food production infrastructure
GM techniques is only about how the seeds are produced. Scale will require mechanisation and automation. Atleast to the extent of what is required as of today. As of today, most of the farm mechanisation equipment in the world is powered by diesel. There is an unsolved, gray-swan waiting to cast its shadow in the name of farm machinery. As long as I don't see an alternative to the ICE deployed in scale, food production is rather going to control populations - not the other way around.
3. Thin margin of error
We all see and hear about Shit Happening all the time. The problem with GM foods is:
1. No ideal "test" environment: Philosophically speaking, you'd need exactly another earth to simulate all the interactions in nature. Flood, drought, wind, sun, water, ecosystem, pollution... just way too many variables in the real environment. The laboratory is but a reduced view of the world. Nature is a complex system and reductionism misses synergetic emergent reactions / behaviors. Worse, experiments are carried out in a production system. As a production operations guy, I completely disapprove this ;)
2. Self replication: Theoretically, there is NO way GM can prevent a chance mutation from helping the GMO in overcoming Terminator technology and being able to reproduce. Once it multiplies, it grows.. and exponentially if the change is to it's advantage! For all you know, maybe that's why they built the Doomsday Seed Vault?
Joseph Tainter, in his book Collapse of Complex societies, argues that Societies exist to solve problems. Given that the theoretical number of problems that one can be challenged with is infinite, there is a point when societies can no longer process information to solve problems. To me, it seems like our Energy predicament will cut down our ability to sustain this information processing ability. Many a times, irreducible complexity is achieved by means of division of labour. However, that means, everybody stops being a generalist. So when one fails (ex: food production), everything fails quickly.
Whether GMO is good for health or not, I can't say. But it sure looks like a recipe for Collapse given the high risk and the nature of problems being tackled!
The fallacy behind the GM argument is due to ourselves - it stems from our very nature - our inflated egos and the over confidence in our ability to solve problems.