Lots of hype but sadly very little science going on here. They have a few superficial models of hippocampal circuitry but almost nothing about how that circuit fits into the larger systems of learning and memory.

There so many major technical issues that make this little more than vapor wear.

The hippocampal neural code is largely un-understood. Without the code they'll be flipping random bits and hoping for the best (akin to flipping bits in ram without understanding how the OS is representing data in RAM).

We still don't have reliable hardware for reading the activity of the brain. Most implants go bad after a few weeks with very few lasting up to a few years. Surgery every few years is unreasonable as each implant will leave behind permanent damage.

Next writing information to the brain is still in its infancy. We can disrupt processing with electrical stimulation but too much current will trigger a seizure (especially in the hippocampus). There's a lot of hope around Opto-genetics but we still need to figure out how to get targeted expression in an already developed human brain. It can take years to get expression right in developing mice.