You really should read these articles before posting them.
1. They admitted that actual studies do not find a link: According to the HLDI, past researchers haven't been able to "definitively connect marijuana use with real-world crashes," and even a federal study failed to find such a link. "Studies on the effects of legalizing marijuana for medical use have also been inconclusive," said the HLDI.
2. They did not do an actual study to look for causation. They simply compared states with legalized pot to their neighboring states. They did not compare them to other states one would expect to see the same effect IF it were true that decreased criminalization causes more accidents, like the very many states with decriminalization. They also did not compare them to states with similar urban vs. rural populations, or similar speed limits, or any other attempt to find a good comparison. They just looked at neighboring states, apparently.
3. As usual, they talk about marijuana "showing up more frequently among people involved in crashes." Well, guess what: marijuana is detectable in urine up to 28 DAYS after smoking, and longer in the blood. I wouldn't be surprised if more people in general had detectable THC metabolites in legalization states, but that doesn't tell us anything about whether they were high at the time or whether their being high was the cause of the accident.
4.They also don't mention whether people are being tested more often after crashes. If police wanted to argue that pot was bad, they would ramp up testing of drivers post-legalization, which in turn would turn up more people who had smoked. That would make the data bad. The rate of testing would have to remain the same for an increase to have a shot at being meaningful.
5. The same results would appear if the unnamed "neighboring states" saw an overall decrease in crashes while in-state crashes remained constant. Because, again, they're talking about collision claim rates (not crash rates, actually) in legalization states relative to collision claim rates in unspecified neighboring states.
6. There also is no comparison to each state's historical rate of crashes. Hell, as noted, we don't even know if the legalization states' crash rates themselves increased. We don't know if there has been a general upward or downward trend lately.
7. As your article notes, the group is funded by insurers whose "vested interested" is "not having to pay claims". Gee, I wonder if they have anything to gain by setting the groundwork for denying claims if someone tests positive for a drug they may not have ingested for four weeks.
8. There is absolutely no discussion of any attempt to determine any other possible causes. Were speed limits on highways raised? Were city speed limits raised? Are more people texting and driving? Etc.
Just because the media says something is "tied to" something does not make it so. The media says all sorts of garbage. You need to carefully parse the article to see what is actually being said, which is that there were more insurance claims submitted for collisions in four legalization states, as compared to neighboring states which may or may not be good comparisons.