AGL 40.00 No Change ▼ 0.00 (0%)
AIRLINK 129.06 Decreased By ▼ -0.47 (-0.36%)
BOP 6.75 Increased By ▲ 0.07 (1.05%)
CNERGY 4.49 Decreased By ▼ -0.14 (-3.02%)
DCL 8.55 Decreased By ▼ -0.39 (-4.36%)
DFML 40.82 Decreased By ▼ -0.87 (-2.09%)
DGKC 80.96 Decreased By ▼ -2.81 (-3.35%)
FCCL 32.77 No Change ▼ 0.00 (0%)
FFBL 74.43 Decreased By ▼ -1.04 (-1.38%)
FFL 11.74 Increased By ▲ 0.27 (2.35%)
HUBC 109.58 Decreased By ▼ -0.97 (-0.88%)
HUMNL 13.75 Decreased By ▼ -0.81 (-5.56%)
KEL 5.31 Decreased By ▼ -0.08 (-1.48%)
KOSM 7.72 Decreased By ▼ -0.68 (-8.1%)
MLCF 38.60 Decreased By ▼ -1.19 (-2.99%)
NBP 63.51 Increased By ▲ 3.22 (5.34%)
OGDC 194.69 Decreased By ▼ -4.97 (-2.49%)
PAEL 25.71 Decreased By ▼ -0.94 (-3.53%)
PIBTL 7.39 Decreased By ▼ -0.27 (-3.52%)
PPL 155.45 Decreased By ▼ -2.47 (-1.56%)
PRL 25.79 Decreased By ▼ -0.94 (-3.52%)
PTC 17.50 Decreased By ▼ -0.96 (-5.2%)
SEARL 78.65 Decreased By ▼ -3.79 (-4.6%)
TELE 7.86 Decreased By ▼ -0.45 (-5.42%)
TOMCL 33.73 Decreased By ▼ -0.78 (-2.26%)
TPLP 8.40 Decreased By ▼ -0.66 (-7.28%)
TREET 16.27 Decreased By ▼ -1.20 (-6.87%)
TRG 58.22 Decreased By ▼ -3.10 (-5.06%)
UNITY 27.49 Increased By ▲ 0.06 (0.22%)
WTL 1.39 Increased By ▲ 0.01 (0.72%)
BR100 10,445 Increased By 38.5 (0.37%)
BR30 31,189 Decreased By -523.9 (-1.65%)
KSE100 97,798 Increased By 469.8 (0.48%)
KSE30 30,481 Increased By 288.3 (0.95%)
Technology

Similar to humans, robots can also develop prejudices

We might be wrong to think that only humans can hold biases and prejudices against outsiders. New research suggests
Published September 7, 2018

We might be wrong to think that only humans can hold biases and prejudices against outsiders. New research suggests that artificial intelligence robots can do that too.

Researchers from Massachusetts Institute of Technology and Cardiff University have found out that similar to humans, AI robots can also develop prejudices while working together by learning from each other.

The study showed that AI can form prejudices by itself. Researchers quoted that ‘groups of autonomous machines could demonstrate prejudice by simply identifying, copying and learning this behavior from one another’, reported Engadget.

In order to test this, a simulated game was set up where each AI robot chose to donate to another within its own group or another group, depending on the reputation of each robot and donation strategies. It was discovered that robots became highly prejudiced against those from other groups.

Artificial Intelligence outshines doctors at predicting heart disease deaths

More than thousands of simulations, the bots learned new strategies by copying one another either within their own group or by across the entire population. The robots copied the strategies that gave them a better payoff in short term, signifying that increased cognitive ability isn’t necessarily needed to construct prejudices.

Co-author Roger Whitaker expressed, “Our simulations show that prejudice is a powerful force of nature and through evolution, it can easily become incentivized in virtual populations, to the detriment of wider connectivity with others. Protection from prejudicial groups can inadvertently lead to individuals forming further prejudicial groups, resulting in a fractured population. Such widespread prejudice is hard to reverse.”

“It is feasible that autonomous machines with the ability to identify with discrimination and copy others could in future be susceptible to prejudicial phenomena that we see in the human population,” he continued.

Moreover, the prejudice fell when there were ‘more distinct subpopulations being present within a population,’ a consideration that is present in human prejudices too, as per Tech Crunch.

“With a greater number of subpopulations, alliances of non-prejudicial groups can cooperate without being exploited. This also diminishes their status as a minority, reducing the susceptibility to prejudice taking hold. However, this also requires circumstances where agents have a higher disposition towards interacting outside of their group,” he continued.

Copyright Business Recorder, 2018

Comments

Comments are closed.