This is something I have been doing in the last couple of years, initially as part of an experiment, but later simply because I enjoy doing it and I am sometimes able to get some photos in the process.
Please note that everything described here is for reference only, and should not be attempted unless you are extremely familiar with the behaviour of hornets. I should note that I generally use grasshoppers purchased from bird shops (many bird shops in Asia sell live grasshoppers for caged birds) for a couple of reasons. For one, I keep some exotic invertebrates such as tarantulas, scorpions and centipedes, so I sometimes feed them with grasshoppers. But more importantly, it is better than to catch grasshoppers from the wild. I do this only if I am out of grasshoppers from the shop. I also sometimes catch flies for this purpose; there are plenty of them around and they can be a nuisance. In short, I do not use anything besides these common insects.
On the first day of 2005, I saw a worker of Vespa bicolor waiting quietly in a clearing; her stillness and alert manner suggested she was waiting to ambush any passing insect. On an impulse, I grabbed a small grasshopper from a nearby patch of grass, tied a length of fine thread to it and lowered it into the clearing, upon which the wasp immediately pounced and started to kill it. What surprised me most was the way in which she first bit the thread off, before continuing to process the prey. I managed to get a few shots before she flew off. As a side note, the daytime temperature on that day was less than 7 degrees Celsius. However, workers of Vespa bicolor were extremely active and busily foraging.
Above: The wasp seizes the grasshopper and attempts to fly off, but the weight of the prey and the attached thread makes it difficult.
Below: She starts biting the thread; it breaks and falls off after mere seconds.
Later, when in Singapore during May-July 2005, I found groups of Vespa affinis foraging among rotting mussels washed up on the beach. I returned to this place several days later with a few mealworms (from an aquarium shop). I tied a length of fishing line to a mealworm and dangled it in front of a foraging hornet. She grabbed it and proceeded to try to bite the fishing line off. However, she was not able to get rid of the comparatively tough line, so she bit the mealworm in half and carried it off. She later returned and took the other half, fishing line and all!
I returned to Hong Kong in July 2005, and it was already summer. Lots of hornets could be seen in the countryside. They would visit the flowers of the banana tree, as well as various types of fruit that fell upon ripening. During this period, although they were eating the fruit to nourish themselves, I found that they would readily grab and proceed to kill any suitable insect which came into contact with them while they were feeding.
I brought some grasshoppers with me whenever I went to such areas. My first attempt was with an individual of Vespa affinis feeding on Sapodilla fruit. This time, I did not use anything to tie the grasshopper. I simply held it by the wings and slowly held it in front of the hornet. She looked up, and the minute she saw and sensed what it was, she pounced and immediately killed it.
I have since done this many times. I have fed workers of Vespa affinis, Vespa velutina and Vespa soror in the same way. Vespa soror is particularly aggressive and voracious in its response to prey offered in this manner.
In mid-winter, it is common to see Vespa bicolor workers perched atop piles of dog droppings, waiting for flies to get within striking range. (See also hunting strategies used by wasps). I caught a few flies with a net, and, braving the stench of the dung, offered the flies to the wasps. They took the flies readily. A video of this can be seen here.
In the summer of 2006, I resumed feeding the hornets whenever I had free time. The year is not over yet, and I shall update only in winter, with whatever new observations I have on this subject. Meanwhile, I will occasionally post new photos of such feedings here.
I offered a worker Vespa velutina a cricket while she was feeding from banana flowers. (I always have crickets at home to feed my exotic pets). She readily grabbed it on the spot, before taking off with it and landing on a higher leaf to kill and process it. In this photo, she can be seen gripping it while a large Vespa tropica continues to feed on the flowers.
I offered another cricket to an individual of Parapolybia indica as well. Again, she was only too pleased with my "offering".
So far, I have offered prey mainly to the hornets (Vespa) and the Parapolybia (lesser paper wasps). This is because the paper wasps (Polistes) prey mainly on caterpillars and usually ignore other insects. I once offered a grasshopper to the giant paper wasp Polistes gigas, but the wasp showed no interest. One species which appears to be an exception is the common paper wasp, Polistes olivaceus. I have seen them catch butterflies on quite a few occasions, and even saw one take down a dragonfly. I will probably try offering a small grasshopper to this species in future.
As a side note, Vespa ducalis and Vespa tropica are two closely related species which are specialist predators on the nests and larvae of Polistines (paper wasps) and Stenogastrines (hover wasps). And I observed that both these species have yet to accept a grasshopper, mealworm or fly from me. They treat such insects as an annoyance, lightly nipping them with their mandibles and then flying away when they find they cannot drive the intruder away! The largest local hornet Vespa soror is a close relative of the well-known Vespa mandarinia, and thus has no such limitations; it is one of the most ferocious insect predators locally and will attack anything it can find; it is thus one of the most exciting species to offer prey to!
I do not think this harms the ecosystem in any way. As I mentioned, I prefer to use crickets, grasshoppers and mealworms which I purchased from elsewhere, and these are plentiful anyway. In the case of the flies, there are literally thousands of them during their peak season, and it certainly does no harm to catch a couple for this purpose.
Is it dangerous to do so? Actually, I offered the grasshoppers to the hornets by hand in the beginning, but have since switched to using forceps. I noticed that three of the regularly-seen species, namely Vespa affinis, Vespa velutina and Vespa soror, sometimes sting their prey, particularly large and strong insects such as grasshoppers. While their aim has so far been extremely accurate, I would not want to be mistakenly stung! Of the five species I have experience with, I still offer prey by hand to Vespa bicolor, because for reasons unknown, I have never seen this species sting its prey. Still, I have to emphasize yet again that this article is only provided for reference.