Now that General Motors’ $2.9-billion second-quarter earnings blew away Wall Street’s expectations, coming up with an encore won’t be easy.

Britain’s departure from the European Union will slow an impressive recovery on the continent by about $400 million in the second half of 2016, the company said Thursday.

Yes, CEO Mary Barra has acted boldly to invest in self-driving cars, ride-sharing and car-sharing, but impatient investors will want to know when and how much those bets will pay off.

This is not to suggest business is about to fall off a cliff. The second half will be very profitable for GM and most of its competitors.

The traditional auto business is in a sweet spot right now, especially in North America. Gas prices remain low. Credit is still easy. Consumers are willing to spend between $35,000 and $70,000 on new trucks, SUV or luxury car. What can go wrong?

But no one yet has figured out how to make money from this wild new world of self-driving cars, ride-sharing and live streaming entertainment into vehicles. Just this year GM has spent $500 million for 9% of Lyft, and $581 million for all of Cruise Automation.

These are proactive moves, but they are not low risk.

The promise, even the illusion, of growth may be worth billions in the tech world, but GM and traditional manufacturers will always be judged by how much money they make now and in the immediate future.

No one understands the need to straddle these old and new worlds of transportation better than Barra. The underlying premise is that GM’s core business will remain very profitable for long enough to enable these new mobility ventures to mature.

“Launching the Chevrolet Bolt EV, getting that into the marketplace and seeing customers’ reactions is one way we can capitalize on our battery cell technology,” she said Thursday morning in response to an analyst’s question about how to monetize new technologies and services. “That’s one of the reasons we did the Lyft alliance, to get it into a ride-sharing fleet.”

She went on to talk about the need to draw more revenue from OnStar, the company’s 21-year-old subscription service that provides security, information, navigation and diagnostic features.

Lyft is not profitable. Cruise is a three-year-old San Francisco startup that never has had to report financial results.

While GM and Cruise employees are testing the latter’s self-driving technology in the Bay Area, they are doing it with the still-under-development Chevrolet Bolt EV, a car that the public won’t be able to buy until the end of this year or early 2017.

With gas prices averaging well below $2.30 a gallon nationally, according to AAA, the appetite for hybrids and plug-in electric cars is tepid at best. At a stated price of $30,000 after federal tax credits, the Bolt will not contribute to GM’s bottom line as much as a $45,000 GMC Sierra pickup or $65,000 Cadillac Escalade.

The good news is that GM’s engineering and manufacturing prowess position it well to lead once these regulatory issues are clarified. It could become the source of vehicles and technology, not just to ride-sharing providers, but to some of its traditional competitors.

Despite the record post-bankruptcy profit GM posted for the April-through-June quarter, its shares rose a modest 1.7 to $32.03 Thursday, still almost a dollar below the $33 price of its November 2010 initial public offering.

GM will try to minimize the impact of last month’s Brexit vote. Does it close either or both of its U.K. assembly plants?

Chuck Stevens, chief financial officer, said it already ships more vehicles from Europe to the U.K. than vice versa. But with the pound sterling falling relative to the euro and U.S. dollar, that means not only are sales in the U.K. likely to fall, but they will be worth fewer dollars.

“Everything’s on the table as we see how this plays out,” Stevens said.

But for the three months ended June 30, the numbers played out pretty well.