An army of ant-sized robots, machines the width of human hairs and custom molecules bouncing through your body - the future is tiny.

An army of tiny, robotic ants stands motionless before a flickering beam of light, waiting patiently for instructions. Within minutes the 120-strong swarm sets off to fetch and retrieve objects as told, negotiating obstacles in their path as the robo-ants constantly keep track of themselves and each other until their work is done.

Welcome to the world of miniature robotics, where the next generation of tiny, smart devices is being dreamt up and groomed to take on the dull, dirty and downright dangerous jobs of the future.

These miniature workers will be honed to perform a host of tasks. Self-powered, mobile and tuned into their environment, they can wriggle into inaccessible places and report back to base.

And whether it's scurrying off to clean the insides of pipes, hovering around the globe to detect environmental change or swimming inside your body to zap disease, micro- and nano-bots are firing imaginations and bringing powerful technologies together.

But at what cost? While experts dispute the science-fiction accounts of self-assembling nanobots ganging up on us and taking over the earth, there are genuine privacy concerns over where tiny monitoring devices could end up, and that the data they generate could be abused.

That doesn't bother the robo-ants though, who have little space for higher intelligence on their four-millimetre-cubed bodies. Equipped with location sensors, an energy source and an onboard computer, each ant is independent and communicates via infrared with its fellow ants in the swarm, explains Dr Ramon Estaña, a researcher with the Institute for Process Control and Robotics at the University of Karlsruhe in Germany.

The robotic ants are the product of I-swarm, a five-year, pan-European project that has gone back to first principles to recreate swarm behaviour in groups of robots. Biologists studied real ants and informed the software programmers, who faced the challenge of developing complex algorithms for the tiny individuals to act like their living cousins, explains Estaña.

And now that the proof-of-principle model is up and swarming, Estaña believes the tiny robotic insects will find niches in industry over the coming decade. "What we are doing now is basic research but we have visions of cleaning surfaces, going into regions where a normal robot can't move," he says.

"At the moment we do not have the space on the robots but in principle it's possible that they can have sensors to smell, and we have one or two robots which can pick up very small parts with a vibrating needle and transport it to the host."

However, these robo-ants are clunky giants compared with microbots being developed at Monash University in Australia to swim through the human body. They want to make a self-propelling device the width of two human hairs that can shimmy along bloodvessels and help doctors perform minimally invasive treatments. Once more, nature provides the cue and the scientists are developing a motorised tail to allow the microbot to swim, much like the one used by the bacterium E.coli.

Smaller again are the nanobots, on the scale of billionths of metres. It's at this scale that we are most likely to see a rapid impact, particularly in areas such as diagnosing disease and delivering highly selective drugs, according to Dr Diarmuid O'Brien, executive director of Trinity's Centre for Research on Adaptive Nanostructures and Nanodevices (Crann).

But the artificial nanobots that course through our blood vessels in the future are not likely to be miniature versions of conventional robots.

At this level it's more about souped-up molecules built from the bottom up by putting chemicals together to create a new structure. Designer molecules are not how many of us would conventionally think of robots, admits O'Brien, but these complex chemicals can be developed to perform highly specific tasks such as delivering therapies in a smart way by locating a target, communicating its position and carrying out an action.

"Typically you would have a gold nanoparticle that you would functionalise with a protein receptor to recognise a particular cancer cell. So you would release it into the bloodstream and it would find the diseased cell. Then you would use a trigger outside the body to get the gold particle to do something, to kill or mutilate or change the cancer cell."

By assembling bots from specific molecules, scientists are again copying recipes from nature, says O'Brien. "It's taking the best part of traditional medicine and pharmacology and linking it up with some of the possibilities that nanomedicine has provided and creating a smarter and more effective, more efficient, safer drug or diagnostic treatment."

But we are still some time off from seeing nanobot-like drug delivery in use, he adds. "The amount of in vivo testing of this capability has been very limited internationally. Even at this point people are talking about possibilities rather than what's happening on a day-to-day basis in labs or in patient treatment."

The hope is that the nano approach will be a shot in the arm for drug discovery, which has become sluggish over recent years. "In general you see a real slowdown in drug discovery and they are not coming out with the volume and regularity of new treatments as before. There is a sense that this approach with nanotechnology might enable a fresh supply of new ideas, drugs and treatments," says O'Brien.

Tiny robots are also the perfect partners for sensors, and this marriage is set to bring monitoring of both health and the environment to a new level.

Hand-held devices could in future allow us to screen for early stages of disease at home. By washing a drop of blood or saliva over a suite of tiny chemical sensors loaded onto a chip, the device could spot "biomarkers" or signs of trouble and send the results wirelessly to a health-care monitoring service.

Problems can be picked up early and in theory treated before they reach crisis point. Monitors in homes could also detect movement and heartbeat and send out an alert if needed. "We are moving into an aging population base and as people get older and more interdependent, you are creating a smart home to act as a third eye on people," says O'Brien.

Big Brother could swing into action on a wider scale too. Micro- and nanobots loaded with sensors could be deployed to monitor environments around the globe and glean valuable data on climate change on land and in the oceans.

But having sensitive medical data whizzing into databases and the prospect of tiny monitors blinking in the atmosphere are prompting genuine concerns over privacy and data protection, and experts would like to see more public discussion about the social changes that lie ahead as technology develops.

In particular, the military is interested in further miniaturising surveillance devices, notes Prof Dermot Diamond, who directs the National Centre for Sensor Research at Dublin City University.

"At the moment microbots and nanobots are in the realm of hype, but a lot of people are very concerned about how you police this and who owns the information and how is the information stored and used," says Diamond. "You need to have checks and balances in there and make sure that this information, once it's acquired, is used properly and is policed properly. I would take the ethical concerns seriously."

This may seem a giant leap from a bunch of metal ants going about their business, but experts warn of the need to ensure the nascent technology is used ethically."It's people's fear of the unknown because we are going into unknown territory and I'm afraid that the track record of human beings using technology in a positive way is not all that good," says Diamond.

IS THIS A ROBOT RISING?

Nanobots are looming on the horizon, but are they going to take over the world?

Not likely, according to experts.

We may have fanciful notions from science-fiction of tiny rocket-like robots invading our bodies, or even self-replicating "grey goo" destroying all in its path, but neither is realistic.

The Fantastic Voyage-like model of a shrunken vehicle just isn't supported by engineering, according to Dr Diarmuid O'Brien of Trinity's Centre for Research on Adaptive Nanostructures and Nanodevices (Crann).

"At the nanoscale those kinds of micro-mechanical machines really just aren't feasible."

Even the bottom-up approach of self-assembling designer molecules has its limitations, and O'Brien dismisses speculation about self-organising nanobots that can take over.

"Molecules behave in a certain way in environmental settings, and applying that in a broader space isn't really credible," he explains.

Much more likely is that social change will creep in on the back of the developing technologies that extend our reach as a species.

As micro- and nanobots improve our capacity to monitor and exploit the earth, store data and detect and treat disease, issues around privacy and healthcare delivery will come to the fore. Because of this, O'Brien calls for more public debate on the matter. Sticking our heads in the sand is of little use, he says.

"From a technological perspective in the 15-20 year timeframe there's really a possibility to use technology to help you manage and order your life in a very different way," he says.

He reckons society will co-adapt as technology develops over the next two decades.

"For example, 20 years ago no one had a mobile phone, and now we now take it for granted, even though many would argue the work-life balance isn't as good because you are more contactable," he says.

"But society has to choose what to accept."