It's not a question of evil robots, but of robots that can be misused, said the University of Washington (UW) study.
"A lot of attention has been paid to robots becoming more intelligent and turning evil," said co-author Tadayoshi Kohno, a UW assistant professor of computer science and engineering.
"But there is a much greater and more near-term risk, and that's bad people who can use robots to do bad things."
He and his colleagues discovered security weaknesses in three robots currently on the market.
The concerns the researchers uncovered with the wireless robots include the fact that the robots' presence is easily detected by distinctive messages sent over the home's wireless network.
The robots' audio and video streams can be intercepted on the home's wireless network or in some cases captured over the Internet.
The authors also identified scenarios in which a robot might physically harm its owner or the home environment. While the risks today are relatively small, researchers say they believe the risks will become more serious as robots become more widespread.
"These are technologies that are being used in the home," noted Denning, a UW doctoral student in computer science and engineering. "The attacks here are very simple. But the consequences can be quite serious," he adds, according to a UW release.
"In the future people may have multiple robots in the home that are much more capable and sophisticated," Denning added.
"Security and privacy risks with future household robots will likely be more severe, which is why we need to start addressing robot security and privacy today."
They presented the findings last week at the International Conference on Ubiquitous Computing in Orlando.