Published 26 July, 2015; last updated 06 July, 2019

The computing power needed to replicate the human brain’s relevant activities has been estimated by various authors, with answers ranging from 10¹² to 10²⁸ FLOPS.

We have not investigated the brain’s performance in FLOPS in detail, nor substantially reviewed the literature since 2015. This page summarizes others’ estimates that we are aware of, as well as the implications of our investigation into brain performance in TEPS.

Sandberg and Bostrom project the processing required to emulate a human brain at different levels of detail.¹ For the three levels that their workshop participants considered most plausible, their estimates are 10¹⁸, 10²², and 10²⁵ FLOPS.

They also summarize other brain compute estimates, as shown below (we reproduce their Table 10).² We have not reviewed these estimates, and some do not appear superficially credible to us.

Drexler looks at multiple comparisons between narrow AI tasks and neural tasks, and finds that they suggest the ‘basic functional capacity’ of the human brain is less than one petaFLOPS (10¹⁵).³

Among a small number of computers we compared⁴, FLOPS and TEPS seem to vary proportionally, at a rate of around 1.7 GTEPS/TFLOP. We also estimate that the human brain performs around  0.18 – 6.4 * 10¹⁴ TEPS. Thus if the FLOPS:TEPS ratio in brains is similar to that in computers, a brain would perform around 0.9 – 33.7 * 10¹⁶ FLOPS.⁵ We have not investigated how similar this ratio is likely to be.

1. From Sandberg and Bostrom, table 9: Processing demands (emulation only, human brain)(p80):

   • spiking neural network: 10¹⁸ FLOPS (Earliest year, $1 million: commodity computer estimate: 2042, supercomputer estimate: 2019)
   • electrophysiology: 10²² FLOPS (Earliest year, $1 million: commodity computer estimate: 2068, supercomputer estimate: 2033)
   • metabolome: 10²⁵ FLOPS (Earliest year, $1 million: commodity computer estimate: 2087, supercomputer estimate: 2044)
2. See appendix A, Nick Bostrom and Anders Sandberg, “Whole Brain Emulation: A Roadmap,” 2008, 130.
3. “Multiple comparisons between narrow AI tasks and narrow neural tasks concur in suggesting that PFLOP/s computational systems exceed the basic functional capacity of the human brain.”

   K Eric Drexler, “Reframing Superintelligence,” 2019, 182.

4. “The [eight] supercomputers measured here consistently achieve around 1-2 GTEPS per scaled TFLOPS (see Figure 3). The median ratio is 1.9 GTEPS/TFLOPS, the mean is 1.7 GTEPS/TFLOP, and the variance 0.14 GTEPS/TFLOP. ” See Relationship between FLOPS and TEPS here for more details
5. 0.18 – 6.4 * 10¹⁴ TEPS =0.18 – 6.4 * 10⁵ GTEPS =0.18 – 6.4 * 10⁵ GTEPS * 1TFLOPS/1.9GTEPS = 9,000-337,000 TFLOPS = 0.9 – 33.7 * 10¹⁶FLOPS