There's no way to answer these questions through thought experiments, you have to look at the available data. There have been several direct basic income experiments, and many other situations that provide relevant data (Native American casino profit distributions for example).
Obviously we need more experiments, but the data we have shows that unemployment tends to go down not up.
>Take away all those 'entry level jobs' and this will decrease GDP no?
Most of the entry level jobs your talking about are being automated away regardless of whether we have a basic income.
But lets remove automation from the equation and assume no increased automation over the next few decades.
You're ignoring the second order effects of a basic income. People who make less than $12,000 a year will definitely make more money, and people who make less than $50,000 or so (whatever the break even point is) who keep working the same amount will also make more money. People in lower income brackets tend to spend their money directly on consumption--this will directly increase GDP.
The data shows that the increase in GDP due to increased spending among the lower income people who make more money each month will likely offset any decreased spending from people who are making less money each month as a result of a basic income plan (either as a result of working less, or paying higher taxes).